1

�� ��原著論文 Original Paper ��

顧客行動の多様性変数を利用した購買行動の予測Deep Learningを応用した実店舗・Web・モバイルの多面的な分析

Predicting Purchases with Using the Variety of Customer BehaviorsAnalysis of the Purchase History and the Browsing History by Deep Learning

新美 潤一郎Junichiro Niimi

名古屋大学大学院Graduate School of Economics, Nagoya Universityniimi.junichiro@h.mbox.nagoya-u.ac.jp

星野 崇宏Takahiro Hoshino

慶應義塾大学Keio Universityhoshino@econ.keio.ac.jp, http://www001.upp.so-net.ne.jp/bayesian/

keywords: electronic commerce, browsing history, purchase prediction, neural network, deep learning

SummaryNowadays, along with the popularity of E-Commerce, the marketing strategy of retail stores has been more

complicated with O2O or Omni-channel. Therefore, Customer Relationship Management (CRM) is one of the im-portant issue for the retail stores. It can be difficult to predict customers future behavior with the simple quantitiveinformation such as purchase frequency since each customers are widely diversified. Although the company canobtain the variety of customers information from their online activity, the use of access history is still limited. Inthis paper, we defined “the variety of user access patterns” collected from their web browsing history and it showsthe patterns they visit the website. Finally, we verified its effectiveness with developing a DNN model to predictcustomers future behavior.

1. は じ め に

1 ·1 研 究 の 背 景

昨今，実店舗を持つ小売企業では，Amazonや楽天といった通信販売サイトの興隆に対抗して顧客を実店舗と

Web上で相互に送客しあう Online to Offline / Offline toOnline (O2O)の試みや，さらにはオムニチャネル ∗1によ

る実店舗，Webの垣根を超えたマーケティング戦略を実施しており，その戦略策定はますます高度で複雑なもの

となっている．

企業が顧客管理に用いることのできる情報は，従来で

は店舗への訪問の際の購買履歴をポイントカードなどを

用いて継続的に収集することで，顧客の購買間隔や商品

カテゴリ等の情報を得るに限られており，店舗訪問時の

購買関連の情報以外を得ることは困難であった．しかし

ながら，このようなオムニチャネルの進行にともなって

企業が得られる顧客データは非常に多面的になっている．

実店舗・Webサイト・モバイルアプリなどの様々なチャネルから消費者にアプローチするオムニチャネルでは，多

くの場合自社顧客を共通の ID等で紐付けていることから，店舗外での日常生活まで含めて幅広い情報を得られ

∗1 顧客の認知から購買までのプロセスに，実店舗，モバイル，Web サイト等の複数チャネルで複合的に対応すること．

るようになった．

企業が顧客管理を行うにあたって，従来からマーケティ

ング分析の手法として一般的であるセグメンテーション

などを用いて顧客を分類することはもちろん有用であ

る．しかしながら，趣味嗜好が多様化した現代社会にお

いて，購買回数等の量的変数やデモグラフィック情報のみ

に頼っての画一的な予測は難しいことが予想される．実

際に [Dembczynski 08] によれば，Webサイトの訪問における検索，回覧，購買などといった行動パターンはユー

ザによって様々に異なることがわかっている．

様々なデータを得られるようになった一方で，多量か

つ多様なデータを未だ顧客の分析に活用しきれていない

のが現状である．企業が新たに得られるようになった顧

客の店舗訪問時以外の日常生活におけるWeb閲覧やモバイルアプリの使用データなど，今後の顧客分析にあたっ

ては幅広い顧客行動を考慮することが重要になると考え

られる．

また，現在様々な分野で機械学習，殊にDeep Learningを用いた予測モデルの構築や解析例が増えている．一般

的には画像や音声といったパターン認識分野に強いとい

われる Deep Learningであるが，近年ではビジネスへの応用も増加している．例としてNTTコミュニケーション

2 人工知能学会論文誌 32巻 2号 B（2017年）

ズにおいては，Deep Learningを含むAI関連技術の活用による電話応対サービスや，得られた家計簿データを分

析してアドバイスを行う機能，さらには画像認識により

消費者が所有する画像を自動分類するサービスなどが提

供されている [伊藤 16]． マーケティング分野においても今後益々の活用が期待されるDeep Learningについて，より精度の高い分析が求められる．

1 ·2 本研究の目的と構成

本研究では，小売企業が自社顧客の将来的な行動を予

測するにあたって，顧客行動の「多様性」を考慮した場

合にその予測精度が向上する可能性について論じている．

マーケティングにおける顧客関係管理の諸研究では，既

存顧客を維持することは新規顧客の獲得に比べて低コス

トであり，より重要であることが示されている [Blattberg08]．新規顧客獲得の戦略策定やコストを考慮すれば，既存顧客の維持は企業にとって重要な課題である．しかし

ながら，そのためには顧客行動をより詳細に把握し，分

析に用いることが必須となる．本研究では，顧客の購買

などといった店舗内での行動はもちろん，顧客の店舗外

での日常生活に少しでも近い情報としてのWeb閲覧の情報の多様性を用いることで，予測精度の改善につながる

可能性を検証する．

Deep Learning等の機械学習では柔軟な予測が可能である一方，現在の機械学習においてひとつの課題となる

のは特徴選択である．特徴選択では組み合わせ爆発や計

算時間の増大を避けるために，主に予測すべき目的変数

に対して説明力のない／小さい特徴量 ∗2をモデルから除

く．現状ではモデルに与えられた特徴量を活用するにあ

たって，そのまま特徴量ベクトルとして扱うことに加え

て複数の特徴量ベクトルを線形結合して扱うことなどが

行われている．しかしながら，それ以上の複雑な変数に

ついては，あくまで必要な特徴量として分析者により用

意される必要がある．そこで本研究ではこれまでもよく

利用されてきた行動の頻度や総量，平均といった量だけ

ではなく，特に顧客行動の多様性という概念に着目する．

例えば ECサイトであれば，同じ回数サイトを閲覧する顧客であっても，多様なサイトからランディングした顧

客と，同一のサイトからの顧客ではサイト上での購買行

動は大きく異なると予想されるため，多様性の何らかの

指標を特徴量とすることで予測精度が向上すると考えら

れる．

本論文の構成は，本章以降 2章において顧客行動のモデリングや Deep Learningを用いて解析を行っている既存研究に関するレビューを，3章において本研究で着目する顧客行動を多様性変数という形で定義する．そして

4章，5章で実際に行った解析例を示し，最終章にて本研

∗2 画像認識においてはある物体を認識するために必要な特徴(色や形，大きさなど)を指し，ベクトルで記述される．より一般的にはある性質／概念を特徴づける変数．

究の考察と今後の課題を示す．

2. 既 存 研 究

顧客行動のモデリング・予測に関する既存研究では，

ECサイトの場合の多くは分析データとして ClickstreamData [Bucklin 02] が用いられている．Clickstream Dataはユーザの探索行動を詳細に記録していることから，ア

クセス・パターン情報等として数値化するにあたって有

用であるとされる [Kim 10]．また実際に顧客の購買行動の予測にこのようなデータを活用した既存研究は数多く

存在している．[Moe 03, Moe 04, Poel 05など]機械学習の技術を用いた予測モデルは現在様々な分野

に応用されているが，Deep Learningの手法として早くに確立されたのが Convolutional Neural Network（CNN[LeCun 89]）である．現在でも特に画像認識の分野においてこの CNN は高い性能を示しており，先述の [伊藤16]を始めとして多くのビジネス事例でも実際に用いられている．

このような分析モデルは，同様にマーケティング分野

においても顧客行動の予測にあたって頻繁に用いられて

いる．例として [Vieira 15]では，Clickstream Dataから消費者の ECサイト上での行動を抽出し，あるセッションでの購買有無の予測および購買された際の商品の予測

を，ロジスティック回帰，Random Forest，Deep BeliefNetwork (DBN [Bengio 07])等を用いて実施し結果を比較している．また購買された商品の予測を行うにあたっ

て，Clickstream Dataに加えて各商品ページに書かれた詳細の文章をword2vec∗3により特徴量ベクトルに変換することで分析に用いている．

また，機械学習モデルを用いて企業のトランザクション

データ等を学習させ，特定の状況下においての顧客行動

を予測した既存研究には，コンビニエンスストア（CVS）の POS（Point-of-Sales）データを用いて顧客エージェントを発生させ，欠品の状況下での購買シミュレーション

を行った [松村 16]や，ID-POSのトランザクションデータと顧客の生活調査データから商品カテゴリと顧客のラ

イフスタイルに関する潜在クラス変数を生成し，ベイジ

アンネットワークを用いて関係をモデル化することによ

り，季節や顧客のライフスタイル，時間帯等の様々な条

件下でどのような潜在クラスの商品カテゴリが購買され

るのかを明らかにした [石垣 11]がある．また顧客管理における顧客の保持／離脱を予測した研

究には，機械学習の中でも様々な手法が採用されている．

Support Vector Machine (SVM [Boser 92])を用いて予測した研究に [Zhao 05]が，またニューラルネットワークを用いて予測した研究には [Sharma 11]などがある．

∗3 Tomas Mikolovらにより開発された 2層のニューラルネットワークであり，テキストコーパスを入力することで文章の単語を特徴量ベクトルとして扱うことのできるアルゴリズム [Mikolov13]．

顧客行動の多様性変数を利用した購買行動の予測 3

Deep Learningに限らない機械学習全般のマーケティングへの活用事例では，複数の消費者や商品同士の類似性

を計算することで，購買される確率の高い商品を表示し顧

客に推薦する協調フィルタリング (Cooperative Filtering[Goldberg 92])を用いた ECサイトでのレコメンデーション等の研究が行われており，既にWeb上で広く活用されている．このような協調フィルタリングを用いたレコメ

ンデーションに関する既存研究やアルゴリズムについて

は，[Park 12]において詳細にレビューされている．ここまでで取り上げてきた代表的な既存研究の多くで

用いられているのは，顧客行動について測定された頻度

や総量，平均値，代表値といった情報であった．例とし

てはある訪問で閲覧・購買された商品の平均単価や閲覧

された総ページ数，訪問した時間帯等である．本研究で

着目しているような，企業が抱える顧客の行動の多様性

に着目して分析を行った研究には [新美 15] がある．これは顧客の自社Webサイト上での行動をユーザ別アクセス・パターン情報として計測し，その行動の質的な多様さ

を多様性変数として定義して自社／競合他社での購買頻

度の予測に用いた研究である．ニュースサイトや ECサイトでの閲覧頻度や購買回数について，ポアソン回帰分

析による予測や，繰り返しのある継続時間と購買有無の

同時モデリングを試みている．アクセス・パターン情報

の多様性の中でも特にリファラ情報 ∗4の多様性がユーザ

のWeb閲覧の多様さを反映する可能性について検討しており，多様性変数を用いた場合にその予測精度が改善し

たことが示されている．このような研究事例からも，従

来顧客行動のモデル化は予測する変数に合わせたパラメ

トリックな統計モデルを用いて実施されることが多かっ

た．しかしながら，今後ビッグデータ活用の流れととも

にこのような予測モデルの作成が積極的に企業内で行わ

れるようになるとした場合，分布を仮定せずに非線形で

柔軟な予測を行える Deep Learning等のモデルが積極的に取り入れられることも考えられる．また，嗜好性の多

様化や様々なデータに由来する変数を組み合わせて予測

を行う場合に，本研究で実施するような非線形の予測モ

デルの方がより予測精度を高められることも考えられる．

よって，本研究では Deep Learningを用いて小売企業における顧客の将来の購買回数の予測モデルの作成を行う

が，その際に顧客の購買データや Web の閲覧履歴データから顧客行動の多様性を定義し分析に考慮することで，

その予測精度が向上することを確認する．

3. 多様性について

3 ·1 顧客行動の多様性とは

ここで顧客行動の多様性の定義，そしてそれを分析で

考慮する意義について示す．まず顧客行動の多様性とは，

∗4 Web サイトの閲覧において，ある Web ページから別のページヘと推移した際に参照元ページの URL を記録した情報．

表 1 顧客が閲覧した商品カテゴリの集計データ例

商品ジャンル 顧客 A 顧客 B

アクセス総数 100回 100回食器 25.0 % 64.0 %収納用品 25.0 % 16.0 %旅行用品 25.0 % 12.0 %寝具 25.0 % 8.0 %

HHI 2500.0 4560.0

顧客が閲覧／購買した商品ジャンルの幅広さや特定のジャ

ンルへの依存度を見ることによる，ある行動の「多様さ」

を定量化した変数である．例としては，ある顧客が調査

期間中に自社サイトで閲覧した商品カテゴリの多様性が

考えられる．具体的には「1. 特定の要素（商品カテゴリ）にどれほど依存しているか」，「2. 要素数がどれほど多いか」の 2点を数値化しており，その際には [新美 15]と同様に，前者に HHIを，後者に累積パーセントを用いた．

3 ·2 HHI の 定 義

HHI（ハフィンダール・ハーシュマン指数）とは，ある市場の寡占状態の度合いを数値化する経済指標であり，

ある産業に参入している競争各社のシェアの二乗和で表

される [Davis 09]．顧客ごとの購買された商品カテゴリの多様性をHHIを用いて表す場合には，ある顧客 kの購買した nk個の商品カテゴリと，当該顧客の購買全体に占

めるそのカテゴリの割合 si について，

HHIk =

nk∑i=1

si2 (1)

と表すことができる．ある要素のシェアが大きいほどHHIは増加し，単一の要素が完全に独占した際に最大値 10000を取る．値が小さいほど多様な要素に依存していること

を表す．例として，自社 ECサイト上で商品ページを 100回閲覧した顧客 A／ Bについて，閲覧した商品カテゴリの内訳を集計した表 1のようなデータが得られたと仮定する．従来ではこのデータから「各カテゴリの閲覧回

数」や「最も頻繁に閲覧された商品カテゴリ」といった

変数を得ることができる．ここでHHIを用いて商品カテゴリについての多様性変数を定義すると，カテゴリごと

のシェアが平等である顧客 Aで HHIが 2500.0，食器にシェアが集中している顧客 Bでは 4560.0となる．特定の要素に依存している場合にHHIが上昇し，アクセス総数が同じであり閲覧した商品カテゴリの数も同じである

ような 2人の顧客について閲覧パターンをより詳細に分離することができた．このHHIについては，実際の解析ではモデルで扱いやすいよう値を取りうる最大値である

10000で除し，0から 1の値として投入している．

4 人工知能学会論文誌 32巻 2号 B（2017年）

3 ·3 累積パーセントの定義

次に累積パーセントについては，あるカテゴリの要素

数の多さを顧客全体の中で相対的に表した値である．購

買した商品カテゴリ数の多様性を例に挙げると，ある顧

客 kの購買した商品のカテゴリ数 nkを全顧客について算

出して昇順に並べた際に，顧客 kが下位何パーセントに位置するかを表した変数である．従って最も要素数の多

かった顧客が 100%をとる．これはある要素の多様性を計測する際に，顧客によっては極端に多くの要素を取っ

ている場合があるが，このような計測方法を用いて他の

顧客と比較した相対的な要素数を用いることで異常値の

影響を除き，正規分布に近づけることを目的としている．

3 ·4 多様性を考慮する意義

このような多様性を考慮することの意義は，自社内で

の限られた行動から顧客の嗜好の幅広さや行動パターン

を推定できることにある．従来的な量的変数としては「ど

れほど多くの商品カテゴリを購買したか」，「どれほど多

くの店舗に訪問したか」などが考えられるが，このよう

な変数は購買回数の増加の結果として増加する場合も考

えられるため，顧客ごとにユニークな行動パターンの説

明に必ずしも貢献しない．一方で，本研究で着目する多

様性変数において主に測定するのは「どれほど特定の要

素に依存しているか」という点である．先の例で言えば，

購買の回数が同一の顧客について「どれほど幅広い商品

を購買したか」を見ながら，その幅広さの中で「どれほ

ど特定の商品ジャンルの購買に依存しているか」を考慮

することにより，単なる購買回数のような量的変数以上

の行動パターンの情報を抽出することが可能であると考

えられる．

例として，図 1に累積での購買回数と来店した店舗の多様性の関係を示す．この図からもわかるように，算出さ

れた多様性変数は必ずしも購買回数とは強い相関を持っ

ておらず，同じ購買頻度の顧客でもこの店舗の多様性に

より「いつも同じ店舗に訪れる顧客」や「幅広い店舗に

訪問した経験のある顧客」など，顧客の行動をより詳細

に把握することができる．また，一般的に購買頻度の高

い顧客ほど購買ジャンルも増加する可能性も考えられる

が，同じ購買回数でも購買されている商品カテゴリの多

様さが異なっていることが分かる．

そして，予測モデルを作成する Deep Learningの特徴抽出においても，現状では複数の変数から「どれほど特

定の要素に依存しているか」といった変数を自動で算出

することはできておらず，分析者が定義して変数として

モデルに投入する必要がある．機械学習技術全般に言え

ることだが，特徴選択により予測に関連する特徴量のみ

に限定することが汎化能力の向上に寄与するということ

は，モデルに投入する特徴量によってその結果が大きく

左右される可能性も考えられる．

さらに，顧客行動の多様性に関する変数は，従来的な

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●● ●●

●

●●●

●

●

●●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●

● ●

●

●

●

●

●

●

●●

●

●

●●

● ●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●● ●●

●●●●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

● ●

●●

●

●

●●

●

●

●

●

●

●

●

●

● ●

●●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

● ●

●

●●

●

●●

●

●

●●

●

●

●

●●

●

●●

●●

●

●

●

●

●

●

●

●

●

●●●

●

●

●●

●

●●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

● ●

●

●

●

●

●

● ●

●

●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●●

●

●

●

●

●

●● ●●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

● ●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●●

●●

●

●

●

●

●

●

●

●

●

●●●

●●

●●

●

●

●●

●●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●●

●

●

●●

●

●

●

●●

●

●

●●●

●

●●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●●

●

●

●●●●

●●

●●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●●

●●

●●

●●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●

●●

●

●●

●●●

●●

●

●

●●●

●

●

●

●

●

●

●●

● ●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●●●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

● ●●

●

●

●

●

●

●●●

●

● ●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●●

●

●

●

● ●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●●

●

●●

●

●

●

●

●●●

●

●

●

●●

●●

●

●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●●●●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●●●

●

●

●

●

●

●

● ●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●●●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●●●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●●

●●

●

●

●

●●

●

●

●

●

●

●

●

● ●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●● ●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●●

●

●

●●

●

●●●●

●

●●

●

●

●●

●●

●

●●

●

●

●●

● ●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

● ●●●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●●

●●●

●●

●

●

●

●●

●

●

●●

●

●

● ●

●

●●

●●

●

●

●

●●

●

● ●

●

●

●

●

●

●

●

●●

●

●

●●

●

●●

●

●

●●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●●

●●

●●

●●

●

●

●

●

● ●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

● ●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●●

●

●●

●●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●●

●

●

●●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●●●

●

●

●

●

●

●●

●

●

●●

●

●

●

●●

●●

●●

●

●

●

●

●

●●

●●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●● ●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●●●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●●●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

● ●

●

●

●●●●

●

●

●

●

● ●

●

●

●

●

●

●●●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●●●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●

●

●●

● ●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●●

●

●

●● ●

●

●●

●

● ●

●

●

●

●●

●

●

●●

●●

●

●●

●●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●●●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●● ●

●

●●

●●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●●●

●●●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●●

●

●

● ●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●●

●●● ●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●

●●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●●●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●●

●●

●

●

●●

●

●●

●

●

●

●●

●●●

●

●●

●

●

● ●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●● ●

●

●●

●

●●

●

●●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●●

● ●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●●●

●●

●●

●● ●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●●

●●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●●●

●

●

●

● ●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●● ●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●●●

●●

●

●

●

●

●

●●●● ●

● ●

●

●

●

●

●●

●

●●

●●

●

●●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●●

●

●●

●●●●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●● ●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

● ●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

● ●

●

●

●

●

●●

●

●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

● ●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

● ●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

● ●

●

●●●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●●

● ●

●

●

●

●

●

● ●

●●

●

●●

●

●●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

● ● ●●

●●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

● ●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●●

●

●●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●●●

●●

●●

●

●

●

●

●●

●●

●

●

●

●

●

●●● ●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●●

●●

●

●

●

● ●●

●

●

●

●

●

●

●

●●

●●

●

●●

●●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●●●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●●

●

●

●

●

● ●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●

●●

●

●●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

● ●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●●●

●

●●

●

●●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●

● ●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●

●●●●

●●●

●

● ●

●

●

●

●

●

●

●●●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●

●●

●●

●●●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●●

●

●●●●

●

●

●

●

●

●●

●

●

●●●

●

●

●

●

●● ●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●●

●

●●

●

●●

●●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●●

●

●●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●●

●

●●

●●

●●

●

● ●●

●

●

●

●

●

●●

●

●●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●●

●●

●

●●

●●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●●

●

●●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●●

●●

●

●●

●

●

●

●

●

●●

●

●

●●

●●

●

●

●●

●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

● ●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●●

●

●

●●

●

●

●

●

●●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●●●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●●

●●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●●

●

●

●●

●

●

●●●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●●●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●●

●●

●

●

●

●

●

●

●●

●●

● ●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●●

●

●●●

●

●

●

●●

●

●

●●

●

●

●

● ●

●●

●

●

●●●

●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●●●

●

●

●

●

●●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●● ●

●

●

●

●

●●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●●●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●● ●

●

●

●

●

●

●

●

●●

●●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●●

●

●●

●

●

●●

●●

●

●

● ●●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●

●●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●

●

●●

●●

●

●

●

●

●●

●

●

●

●●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●

● ●

●

●

●●

●

●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●●

●

●

●●

●

●

●●●

●

●

●

●

●

●

●

●

●

●●

● ●

●

●

●

●

●

●

●

●

●●

● ●

●

●

● ●

●

●

●

●●

●●

●●

●●

●●

●

●

●

●

●

●

●●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●●

●

●

●

●●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●●

●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●●

●

●

●

●

●

●

●

●●

●

●●

●

●

●● ●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●●●

●●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

● ●

●

●●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●●

●

●●

●

●●

●

●

●

●

●●

●

●●●●

●

●●

●

●

●

●●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●●

●

●

●

●

●●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

● ●

●

●

●

●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●●

●●

● ●

●

●●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

● ●

●

●●

●●

●

●

●●

●

● ●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

● ●

●

●

●

●

●

●●

●●

●●

●

●●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●●

●●

●

●

●

●

●

●

●● ●

●●

●

●●

●

●

●

●

●

●●

●●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●●

●

●

●●

●

●

●●

●

●●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●●

●●

●

●●

●

●●

●

●

●

●●

● ●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●●

●

●

●

●

●

●

●

●●●

●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

● ●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●

●●●

●●

●

●●

●

●

●

●●

● ●●

●

●●

●

●●●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●●

●●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●● ●●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●●●

●

●●

●●

●

●●

●●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●●●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●●

●

● ●

●

●●

● ●

●

●

●

●

●●●

●

●●●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●●

●●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●●

●

●

●●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

● ●

●●

●●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●●

●

●●

●

●

●

●●

●

●

●

●

●●

●●

● ●

●

●●

●

●

●

●

●

●●●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●● ● ●

●●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●●

●

●●●

●

●

●

●

●

●

●●

●

●

●●●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●●

●

●●

● ●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●●

●●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●●●●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●● ●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

● ●

●

●

●

●

●

●

●

●

●●

●●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●●

●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●●

●

●

●

●●●

●

●

●●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●●

●●●

●

●

●

●

●●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

● ●

●

●

●

●

●●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●●

●

●●

●

●●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

● ●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

● ●

●

●

●

●● ●

●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●●

●

●

●●

●

●

●●

●

●●

●

●●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

● ●

●●

●

●●

●

●●

● ●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

● ●

●●

●

●

●

●

●

●

●

●

●●

●●

●

●●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●●

●

●

●●

●

●●

●●●

●

●

●●

●●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

● ●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●●

●

●

● ●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●●●

●●

● ●●

●●

●

●

●●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●●

●●

●

●●

●

●●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●●●●

●●

●

●

● ●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●●

●

●

●

●

●

●●

●

●

●●

●●

●

●

●

●●

●●

●●

●

●

●

●

●

●

●

● ●●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●●

●

●●

●●●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●●

●

●

●

●●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●●

●●

●●● ●

●●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●● ●● ●

●

● ●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●●●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

● ●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●● ●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

● ●●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●●

●

●

●

●

●●

● ●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●●

●

●●●

●

●

●

●● ●

●

●

●●

●

●

●

●●

●●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●●

●●

●

●

●●

●

●

●

●

●●

●●

●

●

●

●

●●

●●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●

●

●●●● ●●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●● ●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●●

●

●

●

●

● ●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●●

●

●

●

●

● ●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●●

●

●

●

●

●

●●●

●●

●

●●

●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●●

●●

●

●

●

●

●

●

●●●●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●●●

●

●

●

●

●●

●

●

●●

●

●●●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●● ●

●

●

●

●●

●

●●

●

●

●

●●●

●

●●

●●

●

●

●

●

●

●

●●●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●

●

●

●●●

●

●

●

●

●

●

●● ●

●

●●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●●

●

● ●

●

●

● ●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●●●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

● ● ●

●

●

●

●

●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●

●

●●●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

● ●

●

●

●

●

●●

●

●●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

● ●●

●●

●

●

●

●●

●

●

●●

●

● ●

●●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

● ●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

● ●

●●

●

●

●

●●

●

●

●

●●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

● ●

●

●●

●

●

●

●●●

●

●

●●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●● ●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●●

●●●

●●

●

●

●

●

●

●●

●

●

●

●

●●●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●●

●

●

● ●●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●●

●

●

●●

●

●●

●

●●

●

●

●

●

●●

●

●

●

●●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●●

●

●●

● ●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●●

●●

●

●

●

●

●

● ●

●

●●●

●

●

●

●

●●

●

●

●●●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●●

●

●

●

● ●

● ●●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●●●●●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●●●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●●

●●

●

●

●●●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●●

●

●●

●●

●

●

●●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●●

●

●

●

● ●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●●●●

●●●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●●

● ●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●●●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●● ●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●●●

●

●

●

●●●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

● ●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

● ●●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

● ●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●●

●

●

●

●

●●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●●

●

●

●

●●

●

●

●

●●

●

●●

●

●

●●

●

●

●

●

●

●

● ●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

● ●

●

●●

●●

●

●

●

● ●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●●

● ●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●●●

●

●

●●

●●

●●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●●

●●

●●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●●

●

●●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●●

●

●

●●●

●

●

●●

●

●●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●●

●●

● ●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●● ●

●

●

●●

●

●●

●

●●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●●

●

●●

● ●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●●●

● ●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●●

● ●●

●

●

●

●●

●

●

●

●

●

●

●

●●●

●●

●

●●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●●

●

●

●

●

●

●

●

●●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

● ●

●

●

●●

●●

●

●

●

●●

●

●

●●

●● ●

●

●●

●

●●

●

●

●

●

●●

● ●

●

●

●●

● ●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●●

●

●●

●

●

● ●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●●●

●

●

●

●

●●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

● ●

●●

●

●

●

● ●●

●

●

●

●

●

●●●●

●

●●

● ●●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●●

●

●●

●

●●

●●

●

●

●

●

●●●

●●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●●

●●

●

●● ●

●●

●●

●

●

●

●●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●●

●

●

●

●

●

●

●

●●

●

●●●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●●●

●●

●

●

●

●●

●

● ●

●

●

●●●

●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●●

●●●

●

●

●

●●

●

●

● ●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●●

●●●●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●●

●

●●

●

●●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

● ●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●

●●

● ●

●

●

●

●

●

●

●

●● ●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●

●●

●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●●●

●●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●●

● ●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●●●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●●

●

●

●●

●●

●

●

●

●

●

●

●●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●●●

●

●

●●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●●

●

●

●●

●●●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●●

●

●

●

●

●●

●

●●

●●

●●

●●

●

●●

●

●

●●

●

●●●

●

●

●

●

●

●

●

●

●

●

● ● ●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●●

●

●●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●●

●●

●

●

●

●

●●

●●●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●●

●

●●

●●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●● ●●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●●●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

●

●

●●

●●●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●● ●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●●●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

● ●

●

●

●

●

●●

●

●●

●

●

●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●●

●

●●●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

● ●

●

●

●

●

● ●

●

●

●

●

●

●

●●

●●

●

●

●

●●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●●

●●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

● ●

●

●

●

●

●

● ● ●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

● ●●

●

●

●●

●●●

●

●●●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●●

● ●

●

●

●

● ●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●●●●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●●●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

● ●●●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●●●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●

● ●

●

●●●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●●

●

●●

●

●●●

●●

●●

●

●

● ●

●

●

●

●

●

●

●●

●

●

●●

●●

●●

●

● ●●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

● ●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●●

●

● ●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●● ●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●●

●●●

●

●

● ●●

●

●●

●

●

●●

●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●●

●

● ●●

●

●

●

●

●

● ●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●

● ●●

●

●●

●

●

●

●●

●

●

●

●

●

● ●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●●

●

●●●

●●●

●

●●

●●

●●

●

●

●

●●●●

●

●

●

●

●●

●

●

●●

●

●●

●

●

●

●●

●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●●

●

●

●

●

●

●

●

●●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●●

●

●

●

●●●●

●●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●●●

●

●

●

●

●

●

● ●

●

●●

●

●

●

●

●

●●

●

●

●

●

●●●

●●

●

●

●

●●

●●

●●

●

●

●

●

●

●●●

●

●

●

●●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●●●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

● ●●●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●●●

●

●

●

●

●

●

●

●● ●●

●●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

● ●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

● ●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

● ●●

●

●

●●●

●

●

●●

●

●

●

●●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

● ●

●

●

●

●

● ●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●●

●

●

●●

●

●●

●

● ●

●●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

● ●●

●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

● ●●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●●●

●●

●

●●

●

●

●

●

●

●

●

●●●

●●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

● ●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●●●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●

●

●

●●

●●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●●

●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●●

●

●

●

●

●●

●

●●●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●●

●●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

1 2 3 4 5

0.2

0.4

0.6

0.8

1.0

自社での累積購買回数(対数)

購買

した

店舗

の多

様性

(HHI)

●●

●

●●●

●

●

●● ●●●

●

●

●● ●

●

●●

●

●

●●●

●●

●●●●

●

●●

●●

●

●●●

●

●●●

●●

●●●●

●

●

●

●●●

●●

●

●

●

●

●

●

●●

●●●

●

●

●

●

●

●

●●

●

●

●●

●●

●●

●●

●●

●

●

● ●●

●●

●●

●

●

●●

●

●

●●

●

●●●●●

●●

●

●

●

● ●●

●

●

●●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●●

●● ●

●

●●

●

●

●

●●

●

●

●

●

●●

●●

● ●

●●

●

●

●●

●

●

●●

●

● ●●

●●

●●

●

●

●●● ●●

● ●

●●●

●

●

●

●●

●

●●

●

●● ●●

●●

●●

●●

●●

●

●

●● ●

●

●

●●

●

●

●

●

●

●●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●● ●●

●

●

●●

●

●●

●●

●

●

●

●● ●

●●●●●●

●● ●

●

●

●

●●

●● ●●

●

●

●

●●

●●●

●●

●

●

●

●

●

●●

●

●

●●

● ●●●

●

●

●

●●

●

●●●●

●

●

●

●

●

●●●

●●

●

●●

●

●●●

●

●●

●●

● ●●

●●

●

●

●

●●

●

●●

●

● ●●

●

●

●●●

● ●

●

●●

● ● ●

●

●●

●●●●

● ●

●

●●

● ●

● ●

● ●

●

●

●

●

●

●

●●

●

●●

●

●

● ●●●

●

●● ●

●

●

●

●

●

●●

●●●●

●

●

●

●

●

●

●

●

●●

●

●

● ●●●●

●

●

●

● ●●●

●●

● ●●

●● ●

●

● ●●● ●

●

●●

●

●● ●

●

●●

●●

●

●

●●●

●

●

●

●●

●

●

●●●

●

●

●●

●●

●●

●● ●●

●

●

●

●●●

●

●

●●●

●●

●●

●

●

●

●

●

●●●

●●●

●

●

●●

●

●

●

●

●●●

●

●●

●

●

●

●

●

●

●●

●●

●●● ● ●

●●

●

●

●●●

●

●● ●

● ●

●

●

●●

●●

●

●●

● ● ●

● ●●●●

●

●●

●

●

●●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●●

●●●

●●

●●

●

●

●●

●●●

●●●

●

●

● ●●

●

●

●

●

●

●●

●

●●

●●● ●

●●

●

●

●●

●

●

●●

●● ●

●

●

●●

●

●

●●

●●

●●●

●●●●

●

●

●

●●●

●●

●

●●●

●

● ●

●

●●

●

●

●●

●

●

●

●●● ●

●

●

● ●●

●●

●●

●

● ●

●

● ●

●

●

●

●

●

●

●

●●

●●

●

●

●●●

●

●

●

●

●

●●●

●● ●●●●

●

●

●●

●

●●●

●

●

●●

●

●

●

●●●●●●

●●

●●

●

●

●

●●●●

●●

●

●●

●

●

●

●●

●●●

●

●

● ●●●

●●

●

●

●●

●● ●●●

●●

●●●

●

●● ●●●

●

●

●●

●

●

●●

● ●●●

●●● ●

●

●

●●●

●●

●

●

●

●●

●●●

●●

●●●

●

●●●

●

●●●●

●●●

●

●● ●●●●

●●●

●

●●

●●● ● ●

●● ●

●●

●

● ●●●

●

●●●

●●

●

●

●

●●●

●●

●

●

●●

●●●

●

●

●●

●

●●

●

●

●

●

●

●●

● ●

●

●

●

●●●

●

●

●

●● ●

●

●●

●●

●

●●

●

●

●●●●

●

●●

●● ●●

●●

●●

●●

●

●

●

●

●

●

●

●●

●

●

●●

●●

●●●●●

●●●

●

●

●

●

●

●●

●

● ●●

●●● ●

●

●

●●●

●●

●

●

●

●●●

●

●

●

●●●

●

●● ●

●●●

●

●

●

●

●

●●

●

●

●●

●

●

●●●

●

● ●

●

●

●●

●●

●

●● ● ●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●●●

●●●

● ●● ●● ●

●●

●

●

●

●●●

●

●

● ●●

●

●

●

●●

●●●

● ●

●

●

●●●

●

●●● ●

●●

●

●

●

●●● ●●

●

● ●●●

●

●●

●● ●

●

●

●

●●●

●

●●●

●

●

●●

● ●●

●

●

●●

●

●

●

●●

●●

●

●●

●●●●●

●●

●● ●

●

●

●

●●

●

●●●●●●

●

●● ●

●● ●

●●

●

●

●

●

●

●

●

●

●

● ●●●

●

●

● ●●

● ●●●●● ●

●

●

●

●

●● ●

●

●●●

●●●

● ● ●●●

●

●●●

●

●●●

●

●●

●

●

●

●

● ●●

●●

●

●● ●●

●

●

●●

●

●●●●●

●

●

●

● ●

●●

●

● ●

●

● ●

●●

●

●

●

●

●●●

●

●

●●

●

●

●●

●

●

●

●●

●

● ●●●

●●●

●

●

●●●

● ●

●

●

●

●●●

●●●

●●

●●

●

●

●●

●

●

●

●

●

●

●●

●●●

●●

●● ●●

●

●●●●

●

● ●

●●●

●

●

●●● ●

●

●●●●

●

●

●

●●●

●●

●●●●

●●●

●

●

●

●

●

●

●

●●

●●●

●

●

●●●● ●

●

●

●

●●

●

●

●

●

●●●

●●●

●

●

●

●●

● ●

●

●●●●

●

●

●

●

●

●●

●

●

● ●●●

●●

● ●

●

●●

●●

●

●●

●● ●

●●●

●●

●

●

●

●●

●●

●

●

●●

●

●

●

●●

● ●●

●

●

●

●

● ●

●●

●●●

●●●●

●●

●●

●●●

●

●●

●●●

● ●●●

●●●

●● ●●

●

●●

●●

●● ●●

●

●

●

●

●●●●

●

●● ●●

●

●●

●

●

●

●

●●

●●

●

●

●

●

●●

●●

●●●

●

●●

●

●●●

●

●●

●

●

●●

● ●

●

●●●

●

● ●

●●

●

●●

●

●

●●

●●

●

●

●

●

●

●

●

●●● ●

●

●

●●●

●

●●

●

●●

●

●

●

●●

●

●

●●

●

●●

●

●●

●● ●

●

● ●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●●●

●●●

●●

●

●

●

●

●●●● ●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●●

●● ●

● ●●

●

●

●

●●● ●●

●

●●

●●

●

●●

●

● ●●

●

●

●

●

●●

●

●

●

●●

●

●

●●●●●

● ●●

●

●

●

●●●

●● ●●

●

●

●

●

●

●●● ●

●●●●●

●●

●●●● ●

●●

●●● ●●

● ●

●●● ●

●

●

●●●●

●

●

●●

●●

●

●●

●

●

●●

●●

●●

●●

●

●

●

●●

●●●●

●

●

●●●●

●

●●●

●

●

●●

●●

●●

●

●●

●

●

●● ●●

●

●●

●●●●

●

●●

●●

●

●

● ●●

●●

●●●

● ●●

●

●●

●●

●

●●

●

●

●

●● ●

●

●

●●

●●

●●

● ●

●●●

●● ●●●

● ●●

●●

●

●

●

●

●

● ●

●●

●

●

●●●

●

●●●

●●

● ●● ●

●

●

● ●● ●●

●

●

●

●● ●

● ●●

●

●

● ●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●●

●●●●

●

●● ●●

●●

●●

●

●

●●●●

●●

●● ●●

●●

●● ●●

●●●

●

●●

●

●

●●

●

●

● ●●

●●●

●

●●

●

● ●●

●

●

●●

● ●

● ●●

●

●

●

●

●

●

●●●

●

●

●

●

●●●●

●

●

●●●

●

●

●

●

●●

●

●

●

●●●

●

●

●

●

●●

●

●

● ●

●

● ●●●

●

● ●

●

●

●

●

● ●

●●

●

●

●● ●●

●

●●

●

●●

●●

●

●●●●●

●

●●●

●●

●

●●● ●●●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●●●

●●

●

●●●

●

●●

●●●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●● ●●●

●

●●

●

●●

●

●●

●●

●●●

●

●

●

● ●

●●●●

●

●

●●●●●●

●●

●●

●

●●●●

●

●

●●

●●●

●●●

●

●

● ●

●●

● ●●●

●

●

●●

● ● ●

●

●

●●

●

●

●

●●

●●

●

●●

●●

●

●

●●

●

●●

●●

●

●

●●

●●

●●●●●

●●

●

●●●

●

●

●●

●

●

●

●

●

● ●

●

●

●

●

● ●

●

●

●

●

●●

●●● ●

●

●

●

●●

●

●

●

●

●

●●●

●

●●

●

●●

●

●

●

●

●

●

●●

●●●● ●

●

●●

●

●

●

●

● ●

●

●●

●

●

●●●●●

●●●

●●

●

●

●●

●●

●

●

●●

●●

●● ●●

●

●●

●

●●

●

●

●●

●●

●

●●

●

●

●

●

●●

●●●●●

●●

●

●●●●

● ● ●

●

●●

●●

●

●

●

●●●●

●

●

●

● ●

●

●

●●

●

●

●●●

●

●●

●

●

●●● ●

●

●

●

●

●

●

●●

●

●●

●●

●

●●

●● ●●● ●

●●

●

●

●

●

●

●●

●●●

●

●●

●●

●●●●

●

●

●

●

●

●

●

●

●●

●●

● ●●●●●●

●●

●

●●

●

●●●

●

●●

●

●●●

●●

●

●

●

●

●●

●●

●●●●

●

●●

●●●

●

●●

●

●

●

●

●

●●

●●

●

●

●●

●●

● ●●

●●

●

●

●●●

●

●● ●

●

●●

●

●

●

●● ●

●●

●

●

●

● ●●

●●●

●●●

●

●

●

●

●●●

●

●

●●

●

●● ●●

●●

●

● ● ●●

●● ●

●●

●

●

●

●

●

●● ●

●

●●

● ●

●

●●●●●

●●

●

●●

●●●

●●

●

●●

●

●●●

●

●

●●

●

● ● ●

●

●●

●●

●

●

●

●

●

● ●● ●

●

●●

●

●●

●

●

●

●

●● ●●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●●

● ●

●●●● ●

●

●● ●

●●●

●●

●●

●

●●

●●

●

●●●

●

●●●

●

●● ●

●

●

●

●●●●

●●

●●

●

●

●●

●

●●

●●●

●

●

●

●●●●● ●●

●●

●

●●

●

●

●●

●

●

●

●

●●

●●

●

●

●●

●

●

●

●

●

●●

●●●

● ●

●

●

●

●

●

●

●●

●

●●●

●●●●

●

●

●

●

● ●●

●

●

● ●●

●●

●

●●●

●●

●

●

● ●

●

●●●

●

●

●●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

●

●●

●●

●

●●

●●

●●

●

●

●

●

●● ●

●●

●●●

●●

●

●

●●

●

●●●

●

●●●

●

●

●●

●

●

●

●

●

●

●

●

●●●●●

●●

●●

●

●●

●●

●

● ●●

●●●

●

●

●●

●●●●

●

● ●●

●

●●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●●

●●●

●

●●●

●

●●●

●

●●●

●

●●

●

●●

●

●

●

●●

●●● ●● ●

●

●●

●

●●

●● ●

●

●

●●●

●

●●

●

●●

●

●●

●

●

●●●

●

●●

●●

●

●

● ● ●

●

●

●

●●● ●

●

●

●● ●● ●

●●●●

● ●

●

●

● ●●

●●

●●

●

●●

●●

●

●

●●

●●●●

● ●●

●●

●

●●

●

●

●●●

●

● ●

● ●●●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●●

●

●

● ●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

● ●●●●

●●●

●●●●

●

●●

●

●

●

●

●

●

●● ●●

●

●●

●

●●●

●

●●

●

●

●

●●

●●●

● ●●

●●●

●

●●

●

●●●

●

●

●

●●

●

●

●

●

● ●●

●

●●●●

●●

●

● ●

●

●

●

●●

●●

● ●●●

●

●

●

●

●●●

●

●●●●●

●

●●

●

●

●

●●

●●

●

●

●

●●

●

●●●

●●

●●

●

●

●●

●

●

●

●

●

●

●

● ●● ●

●

●

●● ●●

●

●●●

●

●● ●●

●

●

●

●●

●●●

●

●●

●

●●●

●●●

●

●

●

●

●

●●●

● ●

●

●

●●

●●

●

●

●

●

●

●●

●

●

●

●

●

●

●

● ●●●●●●●

●●

●●

● ●

● ●

●●●

●

●

●

●

●

●

●●●

●●●

● ● ●

●

●●●●

●● ●

●●

●●

●●●

●

●

●●●

●●

●●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●●●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●

●●

● ●

●●●

● ●

●●

●

●●

●●●

●

●●

●

●

●

●

●

● ●●

●

● ●●

●

●

●

●●●

●●

● ●

●

●

●

●

●●●

●

●

●●●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●●

●●

●●

●●

●

●

●● ●●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●● ●●

●●

●●

●

●

●

●

●●

●●

●●

●

●●

●

●

●

●●

●●

●

●● ●

●

●

●

●

●

●

●●

●

●

●

●●●

●

●

●●

●

●

●

●

●

●

●

●

●●●●

●

●●

●●

●

●●●●

●

● ●

●

●

●

●

●●

●●●

●●

●

●

●●

●

●●●● ● ●

●● ●●

●

●

●●

●

●

●

●●

●

●

●

●

●● ●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●●

●

●

●

●

●●●

●

●

●●

●

●

●● ●

●●

●

●

●

●

●

●●

●

●●●

●●

●●

●

●●

●●

●

●

●

●● ●●

●

●

●

●

●●●●●

●

●●

●●

●

●●

●

●● ●●●

●

●

●

●●●

●

●

●●

●

●

●

●●

●

●

● ●●●

●

●●

● ●●

●●

●●

●●

●

●●●

●

●●●●● ●●●

●

●●●

●

●●

●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●●

●

●●

●

● ●●●

●

● ●

●

●

●●

●

●●●

●

●

●

●●

●

●●●●

●

●

●●●

●

●●● ●●●

●●

●

●

●●

●●●

●

●●

●

● ●

●

●● ●

●●●

●

●

●

●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●●

●

● ●●● ●●●

●

●

●

●●●

●●

●●●●

●

●●

●●●

●

●●

●●

●

●●

●

●

●

●●●

●

●

●

●

●

●

●●●

●

●●

●●

●

●

●

●●

●

●●●●

●

●

●

●●

●

● ●●●

●

●●●●●

●●

●

●●

● ●●

●●

●

●

●

●●

●●

●●

●

●

●

●

●

●●

●

● ●●

●●●

●

●●

●

●●

●

●●●

●●

●

●

●●

●●

●●

●●

●●

●

● ●

●

●

●

●

●●●

●●

●●

●

●

●●

●● ●

●

●

● ●

●

●●

●●

●●

●

●●

●● ●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●●

●

●●

●

●

●

●

●●

●●

●●

●●●●

●●●●

●

●●

●

●

●

●●

●●

●

●

●●

●

●●●

●

●

●

●

●

●

●●

●●●●

●●

●

●●

●●

●●

●

●

●●●

●

●●

●●

●

● ●

●●●

●●

●

●●● ●

●●●

●

●●●

●

●●

●

●

●●

●●

● ●

●

●● ●●●● ●

●

●●●

● ●

●●

● ●

●

●

●●

●

● ●● ●●●

●●

●

●●●

●

●●

●●

●

●●●

●

●

●

●● ●

●●

●●

●● ●

●

●●

●●

●

●

●

●● ●

● ●●

●

●

●●

●●

●

● ●

●●

● ●●●

●

●●●●

●

●

●

●

●

●● ●

●

●

●

●●●

●

●●

●

●

●

●●

●●

●●

●

●

●

● ●●

●●

● ● ●●

●

●●

● ●

● ●●

●

●

●●●●

●●

●

●

●●

●

●

●

●

●

●

●●

●

●● ●●●

●● ●

●

●

●

●

●

●●●●●

●

●●●

●●

●

●

● ●

●●●

●

●

●

●●

●

●

●

●

●● ●

● ●●●●●●

●

●

●

● ●●●●● ●●

●

●

●● ●●

●● ●● ●

●

●

●● ●

●●

●

●●

●

●

●

●

●●●

●●

●●

● ●●●

●

●

●

●

●●

●

●● ●

● ●

●●

●

●

●●●

●

● ●

●

●●

●

●

●

●

●●●

●

●

●

●

●

●●●●

● ●●

●

●

●

●

●●

●●

●●●

●●

●

●●

●

●●

●

●●●

●● ●●

●

●● ●

●

●

●●

●●●●

●

●●

●●

●●

●●

●●

●

●

●● ●● ●

●

●

●

●

●● ●

●

●

●

●

●

● ● ●● ●

●

●

●

●●●●●

●●

●

●

●

●

●

● ●●

●

●●●

●

●●

●● ●●●

●●●

● ●

●●●

●

●●●

●

●

●●●

●●

●

●

●

● ●●

●

●

●

●

● ●●● ● ●●●

●

● ●

●

●●

●

●

●●

●●

●

●

●

●●●

●

●●

●

●

●

●●

●

●●●●

●

●●

●●

●

●

●

●● ●●

●●

●

●

●

●

●

●

●●●

●

●

●● ●

● ●●

●

● ●●●●

●

●

●●

●●

●

●●

●

●●

●●

● ● ●●●

●

●●

● ●

●

● ●●

●

● ●●

●

●

●

●●

●

●

●

●

●●

●●●

●●●

●

●

●●

●

●●

●●●

●

●● ●●

●●●

●

●●

●●

●

●

●

●

●

●● ●

●

●

●

●

●●●●●

●

●

●●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●

●●

●●●

●

●●

●● ●

●

●●

●●

● ●●

●

●

● ●●

●

●●

●

●

●

●

●●

●●

●●

●

●

●●

●●●●●

●●●

●

●

●

●●●●

●

●

●●●●

●

●

●

●

●

●

●

●

●

●●

●

● ●

●

●●

●●

●

●

●●●●●

●

●●●

●

●

●

●●

●

●●

●

●

●

●●●

●

●●

●●

●

● ●●

●●

●

●●●

● ●

●

●●

●●●

● ●●

● ●●

●

●

●

●

●

●

●

●

●● ●

●

●

●

●●

●●

●●

●

●●●

●

●

●

●

●

●●●

●

●●

●●● ●

●

●

●

●

●●●●

●● ●●●

●

●

● ●●

●

●

●

●

●

●●

●● ●●●

●●

●

●

●

●●

●●●

●

●

●●

●

●●●

●

●●

●●●

●

●

●●● ●●

●

●

●

●

●●

●

●● ●

● ●

●●●●

●●●

●

●●●

●●

●

●●

●

●

●●

●●

●

●

●

●● ●●

●●●

●● ●

●

●

●

●

●●●

●

●

●●

● ●● ● ●●

● ●

● ●

●●

●

●

●

●●

●

●

●●●

●

●

●

●●

●●●

●

●

●●●

●●

●

●

●

●●

●● ●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●

● ●●●

●

●

●

●

●

●

●●

●

●●● ●

●

●●●

●●●●●

●●

●

●●●●

●

●

●

●

●

●●

●

●●● ●

●

●

●

●

●●

●●

●

●

●

●

●

● ●

●

●

●●

●

●●

●●

●

●

●

●

●

● ● ●●

●● ●●●

●

●

●●

● ●

●

● ●●●

●

●●●●

●

●●

●

●●

●●

●

●●

●

●●

●

●

●

●

●●●

●

●●

●●

●

●●

● ●

●

● ●●

●

●

●

●

●●● ●

●

●

●

●

●

●

●

●

●

●

●●●●

●●

●●

●●

●●●●

●

●

●

●

●●

● ●●

●

●

●●● ●●

●

●●

●

●

●●

●

●●

●●

●

●●

●

●● ●●

●

●

●●

●

● ●●

●

●● ●

●● ●●

●●●

●

●

● ●

●●●●

●

● ●●●

●●●

●● ●●

●●

●●

●●

●●

●

●

●

●

●●●

●

●●●

●

● ●

●

●

●●

●

●

●

●

●●●

●

●●

●●

●

●●

●

●

●

●●

●●

●

●●

● ●

●

●●

●

●

●

●

●

●

● ●●

●

●●●

●●●● ● ●

●●

●

●

●●

●

●

●

●

●

● ●

●●

●

●

● ●●●

●

●

●

●

●●●●

● ●●● ●●●

●

●●

● ●

● ●●● ●

●●

●●●

●

●●●

●

● ●

●

●

●●●● ●

●

●

●

●

●

●●●

●● ●

●

●

●●

●

●●

●

●

●

●●

●

●●

●

●

●

●

●●

●●

●

●

●

●

●

●

●

●

●

● ●

●●

●

●

●

●●●

●

●

●

●●

●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●● ●

●● ●

●

●

●

●

●●

●

●

●

●● ●● ●●

●

● ●●

●

●

● ●●

●●

●●●

●

●●

●

●

●

●

●

●

●●●

●●●

●

●

●●●

●

●● ●

●

●●

●

●●

●

●●

●●

●

●

●●●

●

●●

●●●

●●●

●

●

● ●

●

●

●

●

●●●

●

●

●●

●● ●

●●

●

●

● ●●

●

●

●

● ●●

●

●

●

●

● ●

●

●

●

●

●

●

●

●

●

●

● ●●

●

●

●

●●

●

●

●

●

●

●●●

●●●

●●

●

●●

●

●●

●●●

●●

● ●

●

●

●●●●●

●

●●

●●

●●●

●●●

●

●

●

●

●

●

●●●

●●

●●

●

●

●

●●

●●

●●

●

●●

●●●

●●

●●●

●

●

●

●

●

●

●

●

●

● ●

●

●

●●

●

●

●

●●●

●● ●

●

●●●●●●

● ●

●●●● ●

●●

●●●

●

●●

●

●

●

●

●

●

●

●

●

●

●●

● ●●

●

●

●●

●

●

● ●●●

●

●

●●

●●●

●●●

●

●

●

●

●

●

●

●

●

● ● ●●

●

●●

●

●●●

●●

●

●

●

●

●

●●●

●●

●

●●

●

●

● ●●●

●●

●

●

●

●

●

●

●

●●●

●●

●

●●

●

●●●●

●●

●

●

●

●●

●●

●●● ●● ●

●

●●

●● ●●●

●●

●

● ●

●●

●

●

●●●● ●

●

●

●

●●

●

●●● ●●●●●●

●●

●●

●

● ●●

● ●

●

● ●●

●

●

●● ●

●

●●

●

●

●

●

●

●● ●

●●●

●

● ●●

●●

●

●

●

●●

●● ●●

●

●

●●●

●

●

●

●

●●

●

●●

●

●

●●●●●

●

●

●

●

●

●●●

● ●●

●●●

●

●

●●● ●

●

●

●

●

● ●●

●

●● ●●

●●

●●●

●

●

●

●

●

●

● ●

●

●●

●

●

●

●●●●

●

●●

●● ●

●●● ●

●

●

●

●

●●

●

●●●

●

●

●

●●

●●

●●●

●

●●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●● ●●

●●

●●

●

●

●

●

●●

● ● ●

●

●●

●

●

●

●● ●●●

●●●●

●

●●

●

●●●

●

●

● ●●

●

●●

●

●

●●

●

●●

● ●●●●

●

●

●●●

●●

●●

●●

●

●

●●●●

●●

●●●

●

●●

●●

●

●

●

●

●

●●

●

●

●●

●●●

●

●●●●

●

●●

●●●

●●

●●

●

●

●●●

●

●●●

●

●●

●●

●

●

●●

●●

●

●●

●

●●

●

●

●

●●●

●●

●

●●

● ●

●

●

●

●

●●

●●●

●

●

●●●

●

●●

●

●

●

●●

●●

●●●

●

●

●●●

●●

●

●

●

●

●●

●

●

●●

●

●

●●

● ●

●

●

●●●

●

●

●●●●

● ●

●●●

●●

●

●●●●

●

●

●●

●●●

●●

●

●

●

●

● ●●●●

● ●

●●

●●●

●●● ●●●

●●●

●

●

●

●

●

●

●●●

●

●

●

●●●

●

●

● ●

● ●

●●●

●

●

●

●

●● ●● ●

●

●

●

●

●

●

●

●●●

●

●

● ● ●●●

●

●

●●●

●

●●

●

●

●●

●

●

●●

●

●

●

●

●●

●

●

●● ●

●

●

●

●

●●

●● ●●

●

●● ●

●

●

●

●●●

●

●

●

●

●

●

●

●

●●●

●

●●

●

●●

●

●

●●

●

●

●

● ●●● ●

●

●

●●

●

●

●

●

● ●●●●

●●●

●

●

●●

● ●●

●

●

●●

●

●●● ●●●

●● ●

●●●

●●

●

●●

● ●

●

●●●

●●

●●●

●●

●

●

●

●

●

●

●●●●●●

●

●

● ●●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●●

● ●●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●●●

●●

●

●●● ●●

●

●●●

●●●

●●●●

●●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●●●

●●●●

●

●

●●●

●

● ●●●

●●

●●●●

●

●

●

● ●

●

●●●

●

● ●●

● ●●

●●●●

●

●●

● ●

●

●

●●

●

●

●

●●

●

●●

●

●

●●

●

●● ●

●●

●●

●

●●

●

●●●

●

●●●

●●

●

●

●

●

●●●

●●●●

●

●● ●

●

●

●● ●

●

●

●●●

●

●

● ●●

● ●

●

● ●

●

●●

●●

●●

●

●

●●

●

●

●

●●

●

●● ●●

●

●●●

●

●

●

●

●●

●

●●

●

●

●●

●●

●

●●

●●

●

●

● ●

●

●● ●

●●●

●

●

●

●

●●●

●

●

●

●

●● ●●●● ●●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●●

●●

●●

●●

●

●●

●

●●

●●

●●●●

●

●

●●●

●

●●● ●

●

●

● ●

●●●

●●●

●

●● ●

●

●

●

●

●

●● ●

●

●●

●●

●

●

●●

●●●

●●

●

●●

● ●●

●

●

●

● ●

●

●

●

●

●●

●

● ●

●

●

●

●

●● ●

●

●●

●

●

●

●● ● ●

●●

●●

●

●

●

●●

●

●

●●●

●●●●

●

●

●

●

●

●

●

●●

●

●

●

●

●●●●

●

●

●●

●●

●

●

●

●●

● ●●

●

●

●●●

●

●●●

●●●

●

●●

●

●

●

●●●

●

●●

●

●

●

●

●

● ●

●

●●● ●

●

●●●

●●●

●

●

●

●

●

●●

●

●●

● ●●●

●

●

●

●

●●

●●

●

●

●●

●● ●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●

●

●

●●●

●●●

●

●

●● ●

●

●●

●●

●

●

●

●

●●

●●

●

●

●

●

●

●●

●● ●

●

●

●

●●

●

●●

●●

●

●

●

●

●

●

●●

●●● ●

●●

●

●

●●

●● ●

●●●

●

●●

●●

●

●● ●●

●●●

●

●●

●

●

●

●●●●

●

●●

●

●

●

●

●

●●

●●

●●

●

●●

●●

● ●●

●

●●

●

●

●

●●

●●

●● ● ●

● ●

●● ●

●

●

●

●●●

●●

●●● ●● ●

●

●

●

●

●

●●●

●

●

●

●●

● ●

●

●

● ●

●

●●●●

●

●

●●●●

●

● ●●

●

●

●

●

●●●

●●

●

●

●

●●●

●●

●●

●

●

●●●

●

●

●

●

●

●

●●

●

●

●

●●●

●●

● ●●

●●●

● ●●

●

●

●

●

●

●●

●

● ●

●

●●●

●●

●●

●

●

●

●

●

●●

● ●●

●●

●

●

●

●

●●● ●● ●●

●●

●●

●

●

●●● ●

●

●●●● ●

●●

●●●

●●

●●

●●

●●

●●●●●●

●

●●

●● ●●●

●●

●

●●

●●● ●●

●

●●●

●

●● ●●

●

●

●●

●

●●

●●●●●●

●●●

●

●●

●●

●

●●

●

●

●●

●

●●●

●

●●

●●

●

●●

●

●

●●

●

●●●●

●●

●

●●

●● ●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

● ●

●

●

●

●

●

●●●●

●

●

●

●

●●

●

● ●

●

●●

●

● ●●

●●

●

●● ●

●

●●

●●●

●●

●●●

●●

●

●

●●

●

●

● ●●●

●

●

●

●

●●●

● ●●●●

●●●

●●

●

●●

●●

●

● ●

●●● ●

●●

●●

●●

●●

●●

●

●●

●

●

●●

●

●●

●

●●

●

●●●●

●

●

●

●●

●●●

●●

●

●

●●

●

●●

●●●

●●

●

●●

●

●

●●●

●

●●

●●

●

●

●

●●

●

●

●●●

●

●

●●●

●●

●●

●● ●●

●●

●

●

●

●●

●●●

●

●●●●

●

●

●

●

●●

●●

●●●

●

●

●

●

●●●● ●● ●

●

●

●●

●

●

●●

●

●●

● ●

●●●●●

●●

●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●●

● ●

●●●

●●

●●

●

●

●

●●● ●

●

●

●

●●

●●

●

● ●●●●

●

● ●●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●●

●

●

●

●

●●

● ●●●

●● ●●●

● ●●

● ●● ●●●

●

●●

●

●●●

●

●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●●● ●

●

●●

●●

●●●

●

● ●●●●●●

●●

●●●

●

●

●

●●

●●●

●●

● ●

●

●

●

●●

●●

● ●

●●

●

●●

●●

●

●●

● ●

●●

●●●

●●● ●●

● ● ●

●

●●

●●●

●

●●●●

●

●

●

●

●●●

●●

●

●

●

●

●

●

●

● ●

●●

●●

●●

●

●●

●

●●

●●

●●

●

●

●

●●

●●

●

●

● ●

●

●●

●

●●●

●

●

●●

●●

●

●

●

●●●

●● ●

●●

● ●●● ●

●

● ●

●

●

● ●●

●

●

● ●●●

●

●

●

●

●

● ● ●

●● ●● ● ●●

●●

●

●

●

●●●

●●

●

●

●

●

●●

●● ●

●

●

●●

● ●●●●●

●

●

●●●

●

●●

●●

●

●●

●●

● ●

● ●●●

●

●●

●

●●●

●

●●●

●

●

●

●●

●●

● ●

●● ●●

●

●●

●

●

●●●

●

●

●● ●

●●

●

●●

●

●● ●

●

●●

●

●●●

●

●●

●

●

●●

●

●●●

●●

●

●

●

●

●

●●

●●●●

●

●

●

●● ● ●

●

●●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●

●●

●●

● ●●

●●●

●

●●

●

●

●●

●

●

●●●

●

●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●●●●

●●

●

●

●●

●

● ●●●

●

●●●

●

●

●●

●

●

●

●●

●●

●

●

●●

●●● ●●

●

●

●

●

●

●●

●●●

●●

●● ●

●●●

●●

●

●●

●

●

●

● ●●●

●

●

●●

●

●

●● ●●●●●

●●

●●

●●

●

● ●●

●●●

● ●●

●●●

●●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●●

● ●● ●

●

●

●

●●

●●● ●

●

●

●

●

●

●

●

●●

●●

●

● ●

●●

●

●●●

●

●

●

●●●

●

●

●

●

●

●

●

●●

●●●

●

●

●●

●

●

●●●

●

●

● ●

●●●●

●

●●●

●

●

●●●●

●

●●

●

● ●

●

●●

●●

●

●

●●

●●

●

●

●

●

●

●

●●●●●●

●●●

●

●

●

●

● ●

●

●

●

●●

●

●●

● ●

●●

●●

●

●

●

●

●

●

●● ●

●

●●● ●●

●

●

●●

● ●

●

●●●

●

●

●

●

●●

●

●●● ●

●●

●

●

●●

●●

●●

●●

●

●

●

●●

●

●●

●

●

●

●●

●●

● ●

●●●

●

●●●

●

●●●

●●

●

●

●●

●●●

● ●●●

●

●

●●●

●●

●

●●

●

●● ●●

●

●

●●●

● ●

●

●

●

●●

●

●●●

●

●

●

●●

●●

●

●

●●

●

●

●

●

●●

●

●●

●

●

●●●

●●●

●

●●

●

●●

●

●

●

●

●

●

●

●●

● ● ●●●●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●●●

●

●

●

●●

●

●●

●●

●●●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●

●●

●●●

●

●

●● ●

●

●

●●

●●

●

●●

●

●

● ●●

●● ●

●

●●

●●● ●

●●●

●●● ●

●

●

●

●

●

●●● ●● ●

●

●

●

●

●

●

●●●

●

●

●

●●

●

●●

●

●●●

●●●●●

●

●

● ●

●●

●

● ●●●

●●●

● ●

●

●

●●

●

●

●

●

●

●

●

●●

●●

●

●●

●

●●●

●

●

●●

●

●●

●●

●● ●

●

●

●

●●

●

●

●

●

●●●●●

●

● ●●

●●●●

● ● ●●

●

●●●

●

● ●

●

●●●

●

●

●●●●

●●

●●●●

●

●●

●●●

●

●

●

●

● ●

●

● ●●

●

●

●

●●

●

●●

●●

●

●

●

●

●●

●●

●

●●

●● ●●●●

●● ●●●

●

●

●

●●

●

●●

●

●

●

● ●●

●●●●

●

●

●●

●

●●

●

●

●

●

●●

●●●

●

●●

●

●●●● ●●

●

● ●

●●●

●

●

●●

● ●

●●●

●

●●

●●

●

●

●●

●

●

●

●

●

●●●

●

●

●

●●●

●

●●●

●

●●

●

●

●●

●●●

●●

●

●

●●

●

●●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●

●● ●

●●

●

●●

●●●

●●

●

● ● ●●●

●

●

●

●

●

●

●

●● ●

●

●●

●

●●●

●

●

●●

●

●

●

●

●

● ●

●●

● ●●●

●●●

●●●

●

●

●

●●●●●

●●

●●

●

●

●

●

●

●●●

●●

●●

●

●

●

●●

●●

●●

●

●●

●

●●

●

●

● ●●●

●●

●

●●

●

●

●●

●

●

●●●●

●●

●●

●

●

●

●

●●

●●●

●●●

●

●●

●

●●●

●

●

●

●●

●●

●

●

●

●

●

●

●

●●●

●●

●●●●

●

●

●

●

●●

●

●●●

●

●●

●● ●●●

●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●●

●

● ●

●●

●

●●

●●

●●●●

●

●●

●

●

●●

●

●

●

●●●

●

●●

●●●

●

●

●

●●●

●

●●●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

● ●●

●

●

●

●●●

●

●

●

●

●

●

●

● ●

●

●

●● ●

●

●

●●●

●

●

● ●●●

●

●

●

●●

●●

●●

●

●

●

● ●

●●

●●●●

●

●●

●●

●

●● ●●●

●

●●

● ●

●

●

●

●

●

●

●

●●●

● ●

●

●

●

●●●

●

●●

●

●●

●

●●

●

●

●

●

●

● ●

●

●

●

●

●

●●●

●

●

●

●

●●●●

●

●●

●

●

●

●●● ●

●●

●

●

●

●

●●●

●●

●

●● ●

●

●●●

●

●

●

●●

●

●

●●

●

●● ●

●●●

●●●

●

●● ●

●

●

●●

●● ●

●

●

●●

●●

● ●●

●●

●

●

●●●● ●●

●

●

● ●●

●

●

●● ●

●●

●

●

●●

●

●

●●● ●

●●

●

●●

●●

●●

●● ●●

●

●

●●

●●

●● ●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

●

●● ● ●●● ●

●

●

●●

●

●

●●●

●

●●●

●

●

●●

● ●●

●

● ●●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

● ●●

●

●●

●●

●

●●

●

●●●

●

●

● ●●●

●●

●●●

●●

●

●

●

●● ●

●

●

●●

●

●

●

●●

●●

●

●

●●

●●● ●●

●

●

●

●

●●

●●

●

●

●

●

●●

●●

●●

●●

●

●

●

●●

●●

●●● ●

●

●

●

●●

●●

●●

● ●

●

●●●

●

●

●

●

●

●

●●

●

●●

●

●

●●●●

●

●● ●

●●

●

●●●

●●

●

●●● ●●

●

●●

●

●

●

●

●● ●

●●●

●●●

●

●●●

●●

●●

●●● ●

●

●

● ●●

●

●

●●●

●●

● ●

●●

●

●●

●

●

●●

●●

●●

●

●●

●

●●

●●●

●

●●

●● ●●

●●

●●

●

●

●

●●

●

●●●

●

●

●

●●●

●

●● ●●

●

●●●

●

●●

● ●

●

●

●

●●● ●

●

●

●

●●●

●

●●

●

●●

●●

●●

●

●

●●

●

●

●

●

● ●

●

●●

●●

●

●●

●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

● ●

●

●●●

●

●

●●

● ● ●●

●

●●

●●

●●

●

●●●●

●

●●

●●

●

●●

●

●

●

● ●●●

●●

●

● ●

●●

●

●●●●

●

●●

●●

●

●●

●

●

●

●

●●

●●

●●●

●●

●

●●●

●

●

●

●

●

●

●● ●●

●

●●

● ●●

●

●●●

●●

●

●

●●●

●●

●●●

●

●● ● ●●

●

●

●●

●●

●

●●●

●

●●

●

●

●●●●

●●

●

●●●

●

●

●

●●

●

●●

●

●

●●●●

●●●

●●

●●●

●

●

●

●●

●

●

●●

●

●

●

●● ●●●

●

●●

●●

●

●

●●

●

●

●

●

●●

●●

●

●●

●

●

●● ●●●

●

●

●

●●

●●

●

● ●

●

●

●●●

●●

●

●● ●●

● ● ●●

●

●●

●

●

●

●●●

●●●●

●

●●

●

●

●●

●

●●

●

●●

●

●

●

●

●●

●

●●

●●

●

●

●●

●

●

● ●●

●

●●

●

●

●

●

●

●

●

●●

●●

● ●●

●●

●●

●

●

● ● ●

●

●●

●●●

●

●●

●

●

●

●

●● ●● ●

●

● ●

●

●

●

●●

●● ●

●●

●

●

●

●●●● ●

●

●

●

●●

●

●

●

●

●● ●●

●

● ●●

●

●●

●

●

●

●●●

●●●

●● ●●● ●

●

●

● ●

●●

●

●

●●

●

●●

●●

●

●

●●

●

●

●

●

●

●●

●●●●

●●

●● ●

●

● ● ●●

●●

●●

●

●●

●

●●●●

●

●

●

●●

●●●

●●

●

●

●● ●●● ●●● ●

●●

●

●

●

●

●

●●

●

●●●●

●

●

●● ●●

●

●

●●

●

●

●●

●●●

●

●

●●●

●

●

●●

●

●●●

●

●

●●

●●

●

●

●●● ●●

●●

●

●

●●

●

●

●

●

●●

●●

●

●●●

●

●●

●

●

●

●

●●●●

●●

●

●●

●●●●●

●

●

●

●●●

●

●

●

●

●

● ●●●

●

●●

●●

●●

●

●

●

●

●●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●●●●

●

●● ●●

●●

●

●

●

●●

●

●●●●

●

●●

●●

●

●●●●

●●

●●

●●● ● ●●●●

●●

●●

●

● ●

●

●

●●

●

●

●

●●●

●●

● ●

●

●●●

●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●●●

●●●

●●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●

●

●●

●●

●●

●●●●

●

● ●

●

●

●

●●

●

●●

●●●

●

●●●●

●

●

●

●●

●●

●

●

●●●

●

●

●

●

●

●●

●●

●●

●●

●

●

●

●

●●●●

●

●

●

● ●● ●

●

●●●

●

●●

●

●●●

●

●

●

●

● ●

●

●

●

●

●

●●

●

●

●

●

●

●●

●●

●

●

●

● ●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●●

●●

●

●

●●

●●●● ●●

●

●

●●

● ●

●

●

●

●●

●●

●

●

●●

●

●●

●●●

●●

●●

●●

● ●

●

●●

●

●

●

●●●

●●●

●

●●

●

●●●

●

●●

●

●●

●

●●●

●

●●

●

●●●

● ●●

●●

●

●●

●●

●

●

●●

●●

●

●

●

●●

●●

●●

●

●●

●●●

●

●

●●●●●

●

● ●●

●

●●

●

●

●

●●●

●

● ●●

●●

●

●● ●

●

●●●

●●●

●

●

●

●

●●

●●

●

●

●●●

●

●●

●●●

● ●

●● ●●

●

● ●●

●

●

●

●

●

●

●●

●

●●

●●●

●

●●●●●

●

●

●

●

●

●

●

●

●

●

●●●

●●

●

●●

●●

●

●●

●

●●

●●

●●

●●●

●

●

●●

●

●

● ●●●●

●

●

●

●

●

●●

●●

●●

●●

●

●●●●

●

●●●

●

●

● ●

●●

●

●

●

●●●●

●●

● ●●

●●

●

●

●

●

●

●

●

● ●●

●●●

●

●●

●

●

●

●

●

●

●●●

●●●

●

●

●

●

●●

●

●

●●

●●●

●

●

●●

●●

●●

●

●

● ●

●●

●

●

●

●

●●

●

●

● ●

●

●

●●●

●

●●●

●

●

●

●● ●●●

●●●

●

●● ●

●

●●

●

●●

●●●

●

●●●

●

●

●

●● ●●

●

●●

●

●

●

●●●

●●●

●

●●

●

●●

●

●

● ●●

●●

● ●

●

●

● ●

●●

●

●●●

● ●

●

●●

●●● ● ●●

●●●

●●

●

●●●●

●●●

●●●●●

●

●

● ●

●

●● ●

●

●

●●

●

●

●●●

●●

●● ●

●

●

●●●

●

●●

●

● ●●

●●

●

●●

● ●●

●

●

●

●

●

●

●●●

●

●

●●● ●

●●

●●

●

● ●●

●●

●

●

●●

●●

●

●

●

●

● ● ●●●●

●●

●

●

●●

●

●

●●

●

●

●

●

●●●

●

●

●

●

●

●

●

●

●

●● ●

●

●

●

● ●

●

●●●

●

●●●

●●

● ●

●

●

●●●●

●

● ●●●

●

● ●

●

●●

●

●●

●●

●

●

●

●

●●

●

●●●

●

●

●

●●

●

●

●●

●

●

●● ●●●●

● ●●●

●

●

●●

●

●●●●●

●●

●●● ● ●

●

●

●

●

●●

●

●● ●

●●

●●

●●

●

●

●

●

●

●●

●●

●

●●

●●●●

●

●

●●

●

●●

●

●●

●● ●●

●

● ●

●●

●●

●

●

● ●

●

●●

●● ●

●●

●●

●

●

●

●

●

●

●● ●●●●

●

●

●●●

●●

●

●

●

●●

●

●

●●

●●●

●

●●●●

●

●●

●

●

● ●●

●

●

● ●

●●

●

●●

●

●

●●

●

●

●

●

●

●

●●

●

●●●● ●● ●

●

●

●

●

●

●●●

●

●● ●

●●

●

● ●

●

●●

●●

●

●

●

●●

●●

●●●●

●

●

●

●●

●

●

●

●

●●

●●

●●

●

●

●

●●

● ●

● ●●

● ●●

●●●

●●●

●

●●

●●

●●●●

●

● ●

●●

●

● ●● ●●

●

●

● ●

●

●

●●

●

●●

●●

●●

●

●

●

● ●●

●

●

●

●

●

● ●

●

●

●● ●

●●

●●

●

●

●

●

●●

●

●●●

●●●

●

●●●

●

●

●●●

●●●●

●●

●

●●

● ●

●

●

●

●

●●

●●●

●

●

●●

●

●

●

●

●

● ●●

●

●●

●

●

●

●

●

●

● ●

●●●

●

●●●

●● ●●

●●

●●●

●

●

●

●

●●●

●

●

●

●

●

●

●●

●●

●●

●

●●● ●

●

●

●●

●

●

●

●●●●

●

●

●●

●●

●

●●●

●●

●

●●●

●

●

●

●●●●

●

●

●

●●●

●

●● ●●●

●

●

●

●

●●

●●●●●

●

●

●●

●

●

●●

●

● ●●

●●

●●

●●

●

●

●●

●●

●

●

●

●

●●

●●●

●●

●

●

●

●●●

●● ●●

●

●

●●

●●●

●

● ●

●●

●●

●●●

●

●●

●

●●

●

●

●

●●

●●● ●● ●

●

●

●●

●●

●● ●●

●● ●●

● ●

●

●●

●

●

●●● ●

●

●

●

●

●●● ●

●● ●●●

●●

●

●

●

●●● ●● ●

● ●

●●

●

●

●

●

●●● ●●●

●

●

●

●

●

●

●● ●●●

●

●

●

●

●

●

●

●

●

●●

●

●●

●●

●

●●● ●

●

●●

●●● ●● ●●

●

●

●

●

●●

●

● ●●

●

●●●●

● ●●

●

●

●●●

●

●●●

●

●●

●

●

●●

●

●

●

●

●●

●●

●

●●

●●

●●●

●

●● ●●

●

●●●

●●● ●

●● ●●

● ●●●

●

●●

●●

●● ●

●

●●●

●● ●

●

●

●

●●●

●●● ●●●

●

●●

●

●●

●

●● ●

●

●●

●

●●

●● ● ●

●

●

●

●

●

●

●

●

●

●

●●●

●

●●

●●

●

●

●

●

●●●

●

● ●●●

●

●

●●●

●●

●

●

●●

●●

●

●

●

●

●

●●

●

●

● ●

●●●

● ●●

●

●●

●●

●

●

●●

●

●

●●

●

●

●

●

●

●●

●

●

●

●●●

●

●

●

●

●

●●● ●●

●●

●

●

●

●●

● ●

●

● ●● ●●

●

●

●●

●

●

●

●

●●●

●

●

● ●●

●●

●●●

●

●

●

●●●

●●

●● ●●

●

●

●

●●

●

●

●●

●●●

●

●●

●●●

●

●

●●

●

●●

●

●●

●

●

●●

●

●

●

●

●●●

●

●

●● ●●●

●●

●●

●●

●●

●

●● ●

●●●

● ●

●● ●●●

●

●

●

●

●

●

●

●

●

●●●

●

●● ● ●●

●

●

●●

●

●

●

●●

●●

●●

●

●●

●●●●●

●●

●●

●●

●

●

●●●●●

●●

●

●●

●●

●

●

●

●

●

●

●●

●

●

●

●● ●

●●

●

●●

●

●

●●

●

●●●●

●

●

●

●●

●

●

●●

●●

●

●●

●

●

●●●

● ●

● ●●●

●●

●

●

●●●

●●●

●

●

●●

●

●

●● ●●

●

●

●●

●●●

●

●●●

●

●

●●

●●

●

● ●●

●

●

●

●●●

●

●●●

●●● ●

●●

●

●●● ●

●

●

●

●

●●

●●●●

●

●

●

●●●

●

●

●

●

●

●

●

●

●●

●

●●●

●●●

●

●

●

●●

●

●

●●

●

●

●●●

● ● ●●

●●

●

●

●

● ●

●●

●

●

●

●

●●

●

●

●●

●

●● ●●

●●

●●●

●

●

●

●

●

●

●

●●● ●

●●●

●

● ●

●

●●

●●

● ●

●●●

●●

●

●●●

●●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●●●

●

●

●●●

● ●

●

●●

●

●

●

●

●

●● ● ●●

●● ●

●●

●●

●

●

●●

●

●●●

●●

●

●●●

●

●

●●

●●●●

●●

●

●

●

●

●

●

●

●●

●●

●●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●●

●●

●●

●

●●●

●

●

● ●●

●●

●

●●

●

●

●●●

●●●

●●

●

● ●

●

●

●●●

●

●●

●

● ●●●

●

●

●

●

●

●●

●●

● ● ●

●●●●

●●

●

●● ●

●

●●

●

●

●●

●

●●

●●●

●●

●●●

●●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

●

●

●●

●

●●

● ●

●

●

●

●●

●

●

●

● ●●

●

●● ● ●

●

●●●

●

●

●

●●

●

●●●●●

●●●●

●●

●

●

●

●

●● ●

● ●●

●●

●

●

●

●

●

●

●

●

●●●●●

● ●●●●●

●

●●

●●●

●

● ● ●● ●

●

●●

●

●●

●

●● ●●●●

●

●●

●

●

●

●

●

●

●●●

●●

●

●●●● ●

●●

●

●

●

●

●

●●

●●

●●

●●●●

●

●

●●

●

●

●

●●

●

●●

●● ●●

●

●

●●

●

●●

●

●

●

●

●●

●

●●●

● ●

●

●●

●

●●

●

●

●

●

●●●

●

●

●●

●●

●● ●●●●

● ●●

●

●

●●

●

●●●

●

●

●●●

●●●

●●

●●

●

●●●

● ●

●●

●● ● ●●

●●●●

●●

●

●

●

●

●

●●● ●

●

●●

●

●

●

●

●

●

●

●

● ●

●

●

●●

●

●

●

●

●

●●

●

●

●●●

●●●●

●●

●

●

●

●

●●

●●

●●●●

●

●

●

●●●

●●

●●

● ●● ●

●●

●● ●

●

●

●

●●●

●●●

● ●

●

●●

● ●●

●●●

● ●●

●

●●

●

●

●

●

●

●

● ●

●●●

●●

●●●

● ●●

●

●

●●

●● ●●

●● ●

●●

●

●●

●

●

●●

●

●●

●

●

●●

●●

●

●

●●

●

●

● ●●

●

●●

●

●●

●

●

●

●

●

●

●

●●●

●●

●

●●

●

●●●

●

●●

●

●

●

●●

● ●

●

●

●

●

●●

●

●●

●

●●

●●

● ●●

●●● ●

●●

● ●●

●

●●

●●

●●

●

●●● ●●

●●●●

●

●

●

●

● ●●

●

●

●

●

●

●

●●

●

●

●● ●

●

●

● ●●●●

●

●

●●●

●

●●●

●

●●

● ●●

●

●●

●

●

●

●●

●

●●

● ●●●●

●●

●

●

●●●●

●

● ●

●

●●

●●● ●

●

●●●

●●●●

● ●●●

●

●● ●●

●

●●

●

●●

●●●

●●

●●

●●●

●●

●

●●

●

● ●●

● ●●

●

●

● ●●

●

●●

●

●●●

●

●●

●●

●●

●

●

●

●

●

●

●

●

●

● ●

●

●

●

●

●

●●

●

●●

●●●

●

●

●●

●●

●●●

●●●

●●●

●

●

●

●

●

●●

● ●●

●

●●

●

●

●●

●●●● ●

● ●

● ●●

●

●●●●

●●

●

●●●

●

●

●●

●●

●

●

●

●

●

●●

●●

●

●

●

●

●

●●

●●

●●●

●

●● ●

●

●

●

●

●●●

●

●●

●

● ●●

●

●

●

●●

●

●

●

●

●

●

●●

●

●●●

●● ●

●

●●●

●●

●

●

●●

●

●

●●●

●

●●

●

●●

●●●

●●

●

●

●

●●

●●

●

●●

●

●●

●●

●●

●

●

●

●

●

●●

●

●● ● ● ●

●● ●●

●

●

●

●●●

●●

●●

●●●

●

●●●●●●

●●●

●

●

●

●

●●

●● ●●●

●

●

● ●

●●

●

●●

●●

●

●●

●

●●●●

●

●●●●

●

●●

●

●

●

●

●●

●●

●

●

●●

●●●●

●●●●

●

●●

●●

●●

●

●

●

●

●●●

●●

●

●●●

●●

●

●

●●

●

●●● ●●

●

●●

● ●

●

●

●

●

●

●●

●

●●

●

●

●●● ●

●

●

●●

●

●

●

●●

●

●

●

●

●●

●

●

●●

●●● ●

●● ●●

●●

●

●●

●

● ●● ●●●

●

●●●

●

●

●●

●●

●

●

●●

●

●●

●

●●●

●●

●

●●

●

●

●

●

●

●

●

●●●

●●

●●●

●● ●●●●●

●

●●●●●

● ●

●

●●

●

●

●

●

●

●● ●

●

●●●

●

●●●●

●

●●

●

●

●

●

●

● ●

●

●

●●

●

●●●● ●

●

●●

●

●●●

●

●

●

●

●● ●

●●

● ●●

●

●●●●

●●

●

●

●

●●●●

●

●

●

●

●

● ●

●

●

●

●●●

●●

●

●●● ●

●

●

●●●

●●●● ●

●

●●

●

●● ●

●●●

●

●

●

●●●

●

●

●●

●●

●

●

●●

●

●●

●●●

●

● ●

● ●

●

●

●

●

●●●

●●●●

●●

●

●

●

●

●●●●

●●

●

●●●

●

●

●

●●

●●● ●●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●●

●

●●●

●

●●

●

●

●

●

●

●●●

●

●●●●

● ●●

●●

●●●

●

●

●●

●

●●

●

●

●

●●

●●

●

●

●

●

●●●●

●●

●

●

● ●

● ●●

●

●

●● ●

●●● ●●

●

●

●

●● ●

●

●

●●

●

●

●●●

●

●

● ●●

●

●

●

●

●

●

●

●●

●

●

●

●

●

●●

●

●

●●●

●

●●●●●

●●

●

●

●

●

●●

●● ●

●

●

●

●

●

● ●

●

●

● ●

●

●

●

●● ●●

●

●●

●

●

●

●

●

●

●

●●

●

●

●● ●●

●

●●●

●●

●

●●

●

●● ●

●●●

●●●

●

●

●●

●

●●

●

●●

●● ●

●

●

●●

●●●

●

●

●●

●

●

●●● ●

●●●●

●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

●●●

●

●●●

●●

●

●●●

●●

●

●

●

●

●

●

● ●●

●●

●●

●

●●

● ●

●

●●

●

●

●

●

●

●

●

●

●

●●●

●

●●

●●

●● ●

●●

●●●●

●

●● ●

●

●●

●

●

●

●●

●●

●

●

●

●

●

●

● ●●●

●● ●

●

●

●

●● ●

●●●●

● ●●●

● ●

●

●

●

●

●●

●

●●

●

●

●●

●●

●●●

●●●

●●

●

●● ●●

●●

●

●●●

●

●●●●

●

●

●

● ●●

●

●

●

●

●● ●

●●

●●●●

●●

●

●●

●●

●●

●●

●●●

●●●●

●

●

●●●

●

●●

●

●●

●

●●●

●

●

●

●

●●

●

●●●

●

●

●● ●●

●●

●

●

●●●

●●

●● ● ●● ●

●●●

●

●●

●

●

●●

●

●●●

●

●

●●

●

●

●

●●

●

●●

●●

●

●

●●

●

●

● ●●●

●

●●

●●

●

●●

●

●●

● ● ●●●

●

● ●● ●●●

●●

●●●

●

●

●

●●●

●

● ●

● ●

●

●

●

●

●

●

●●

● ●● ●

●

● ●

●

●● ●●

●

●

●

●

●

●

●●●

●

●●

●

●

●●

●

●●

●● ●

●

●

●●

●

●

●●●

●

●

●

●●

● ●

●●●●●

●

●●●

●

●

●

●

●●● ●

●

●

●●

●●

●

●

● ● ●●●● ●

●●

●

●

●●●

●● ●

●

●

●

●

●

●

●●

●●

●

● ●

●●

●

●

●

●●

●

●

● ● ●●

●

●●

●

●

●

●

●

●

●●

● ●

●

●●●

●●

●

●

●●●

●

●

●

●

●●

●

●

●

●

●●●●

●

●

●●●

●●

●

●●●●

●

●

●

●

●

●

●

●

●

●

●

●●●●

●●●●

●

●●●●

●● ●●

●●●

●● ●

●

●

●●●

●●

●

●●

●

●

●●

●●

●●

● ●

●●

●

●

●

●

●

●

●

●●

●●●●●

●●

●

●

●●

●●

●

●● ● ●

●

●

●

●●

●

● ●● ●

●

●●●

●

●●

●

●● ●

●

●●

●●

●● ●

●

●●●

●●

●●●

●●

●●

●●

●

●●●●●

● ●●●●

●

●●●

●

●●●

●●

●

●●●

●

●

●

●●●●●

●● ●●●

●●

●●

●

●

●

●

● ●●

●●

●

●

●

●●

●

●●

●● ●

●

●●●

● ●●

●●

●

● ● ●

● ●● ●●●

●●

●

●●● ●●

●

● ●

●

●

●●

● ●

● ●●

●●●

●●●

●

● ●

●

●

●

● ●●●

●

●●

●

●

●●●

●●

●●

●●●

●●●

●

●

●

●

●

●

●

●

●● ●●

●

●

●●●

●●

●

●

●●

● ●

●●●●●

●

●

●

●

●

●●●

●

●

●

●

●

●

● ●● ●● ●●

●

●●

●

●●

●

●● ●●●

● ●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●●

●

●●●● ●●

●●

● ●

●

●

●

●

●●●

●● ●●

●● ●

●

●●

●

●●

●●

●●

●●●

●●

● ●

●

●

●●●

●

●

●

●●

●

●●

●

●●

●●

●●

●

● ● ●

●

●●

●

●●●

●●

●

●

● ●●

●

●

●

●●

●

●●●

●●●

●

●

●●●

●

●●

●●

●●●

● ●● ●●

●●

●●●

●

●●

●

●●

●

●

●●●

●●● ●●

●

●

●●

●

●●

● ●●

●

●

●

●

●

●

●●●●

●●

●●

●●

●

●●

●●●

●

●

●

●●

●●

●

●● ●

●

●

●

●●●

●●●

●

●

●

●●●

●●

●

●

●

●●

●●●

●

●●

●

●

●

●

●

● ●●●

●

● ●●

●●

●

● ●

●●●

●●

●●

●●

●●

●

●● ●●

●

●●

●

●●

●●

●

●

●

●●

●

●●

●

●●●

● ●●

●

●

●●

●

●●

●●

●

●●

●●

● ●●●

●●

●

●

●

●●

●●

●●●

● ●●●

●●

●●

●

●●

●

●

●

●●●

●

● ● ●

●

●

●

●●●

●●

●

●●

●

●●●● ●●

● ●●●

●●

●●●●

● ●

●●●

●●

●●

●

●

●

●

●

●●

●

●●●

●●

●

●●●

●

●

●●

●●

●●

●

●●

●●

●

●●●

●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●

●

●

● ● ●●●

●

●

●

●●

●●●

●● ●●●●

●

●●

● ●

● ●●

● ●●

●

●●●

●

●

●●

●

●

●●

●

●

●

●●

●●

●● ●

●

●●

●●●

●●

●●

●

●

●●

● ●

●

●

●

●●●●

●

●

●

●●

●

●

●

●

●

●●

●

●●

●●●●

●●

●●

●

●

●

●●

●●

●●

● ●●●

●

●●

●

●●

●

●●

●

●

● ●●

●

●

●

●●

●

●

●●● ●

●

●

●

●

●●

●● ●

●

●

●●

●

●

●

●

●

●●●●

●

●

●●●●

●

●●

●

●●●

●●

●

●

●

●●●

●●

●

●●

●

●●

●

●●

●

● ●

●

●●●●●●

●● ●●

●

●

●

●●

●

●

●

●●

●

●●●

●●

●●

● ●

●

●

●

●●

●

●●

●

●

●

●●

●●

●

●

●

●●

●●

●●●

●

● ●●●

●●

●

●

● ●●●

●

●

●

●

● ●●

●

●

●

●●●●

●

●

●

●●●●

●●●●●●

●

●

●

●

●●

●●

●

●●● ●

●

● ●

●

●●●

●

●●

●●

●●

●

●●

●●●

●●●●●

●

●●

●●

●●

●●

●●

●●

● ●●

●●

● ●

●

● ●

●

●● ●●

●● ●

●

●●

●

●●

●●

●

●

●●

●●●

●●●●

●●

●●●●●●

●●●

●

● ●

●

●

●● ●

●

●

●

●

● ●●●

●

●

●

●●

●

●●

●●

●● ●●

●

●●

●

●

●●

● ●

●●

●

●

●●

●

●

●

●●

●●

●

●

●

●●

● ●

●●

●

●●●●●●

●

● ●

●

●

●●● ●●●

●

●

●●

● ●

● ●●

●●

●

●

●

●● ●●● ●

●

●

●

●

●

●

●●●

●

●

●●

●

●

●●●

●●

●

●●

●

●●●

●●

●●

●

●

●●

●●

●●●

●

●●

●●

●●

●

●

●

●

●●●

●

●●●●●

●

●

●

●

●

●

● ●● ●●

●●●

●●

●

●

●●●●

●● ●●

●

●

●

●

●

●

●

●

●

●●

●

●

●

●●

●

●

●●●

●

●

● ●

●

● ●●

●

●

●

●

●●

●● ●●

●

●●

●

●

●●●

●

●

●

●

●●

●●

●

●

●

●●●●

●

●

●

●

●

●●●

●

●

●●● ●●●●

●● ●●

●

●

●

● ●●

●

●

● ●

●●

●●

●●●●

● ●

●

●

●●

●

●●

●

●●

● ●●●

●

●●

●●

●

●

●

●●

●

●

●●●

● ●●

●●●

●●

●

●● ●

●

●

●

●

●●●

●●

● ●

●

●

●

●

●●●

●

●

●

●●

●●

●●

●

●

●●

●

●

●●● ●●

●●

●●

●

●●

●●

●

● ●●● ●●●

●

●●

●

●

●

●●●●

●

●●

●●

●

●

●●

●

●

●

●

●

●

●●

●●● ●

●

●

●

●●●

●

● ●

●●●

●

●●

●●●●

●

●●●

●● ●

●

●

●

●

●

●

●●

●● ●

● ●

●●

●

●●●●

●

●

●

●

●●●

● ●●

●●

● ●

●

●

●●

●

●●●

●

●

●●

●

● ●

●

●

●●

●

●

●

●●●●

●●

●

●●●

●

●

●

●●●●

● ●●●●

●●

●●

●

●● ●●

●

●●

●

●

●

●

●

●

●

●

●

●

●● ●

●

●

●●●

●

●

●

●●

●

●● ●

●

●●

●● ●

●

●

●●●

● ●●

●

●●●

●

●●● ●

●

●

●

●●

●

●

●

●●

●

●

●

●

●

●●●

●●

●●●

●

●

●●

●

●●●

●

●

●

●

●

●●●

●

●●

●●

●

●

● ●●●●

●

●● ●●

●●●

● ●

● ●

●

●●

●●

● ●

● ●

●

●

●

●●● ●●●

●

●

●●●

●●●

● ●

●

●●

●●●

●

● ●

●

●

●

●

● ●●●

●●

●

● ●

●

●

●

● ●●●

●●●

●

●

●●

●●

●

●● ● ●●

●● ●●●● ●●

●

●●

●

●

●

●

●●

●

●

●

●

●

●

●

●

●

● ●●

●

●

●

●

●

●

●

●●

●

●●●

●●● ●●

●●

●

● ●

●●

●

●

●

●

● ●

●●

●● ●

●

●●

●

●

●●●

●●●

●

●●●●

●

●

●

●

●

● ●

●

●●

●

●●●

●

●●

●

● ●● ●●

●●●●●

●

●

●●

●

●

●

●●●

●

● ●

●

●

●●

●

●●

●●●

●

●●

●

●

●

●

●

●

●

●

●●●

●● ●

●●

●

●

●●

●

●●

●

●●

●

●

●

●●

●

●●●

● ●●

●

●

●●

● ●

●

●

●

●

●

●●●

●

●●

●

●

●●●●

●

●

●

●

●●●●

●

●

●●

●

●●●

● ●

●

●●

●

●●

●

●

●●●

●

●

●●

●

●

●

● ●●●●

● ●

●

●

●

●●

●●

●● ●

●●●

●●

●●●

●

●

●

●

●●

●

●●

●

●●

●●

● ●

●

●

●

●

● ●

●

●

●●●●

●

●●●●●

●

●

●●

●●●

●●

●●●

●

●

●

●

●

●●●

●●

●●●

●

●●●●

●

●

●

●

●●

●

●

●

●

●

●●●

● ●

●

●

●●

●●

●

●

●●

●●

●●● ●

●●

●

●●

●

●

●●

●

●●●

●●

●

●

●

●

●● ●●

●

●

●

●

●

●

●●

● ●● ●●●●

● ●

●

●

●

● ●

●●

●

●

●●

●

●

●

●

●

●●

●●●

●

●●

●

●

●●

●●

●

●

●

●

● ●●

●

●

●

●

●● ●

●

●●●●

●●●●

●

●●●

●

●

●●

●●

●●

●●

●

●

●

●

●

●●

● ●

●

●

●

●

●●

● ●

●

●●● ●● ●

●

●

●

●

●●

●

●●

●●

●

●●

●●

●●

●

●●

●

●●

●

●

●

●●

● ●

●●

●●

●●●

●

●

●

●

●●

●

●

● ●

●●

●

●

●●●

●

●●

●● ●

●

●

●● ●●

●

●●

●

●●

●

● ●

●

●●

●

●

●

●

●

●

●

●●

●●

●●

●●●

●●●●●●

●

●●

●

●

●●

●●

●

●●

●

●●

●

●●●●● ●

●

●●●

●●

●●●

●

●

●

●

●● ●

●

●●

●

●

●● ●

●

● ●● ●●

●●

●●

● ●●●

●●●

●

●●

●

●

●

● ●●●

●●

●

●

●

●

●

●

●●

●●

●

●

●

●●

●

●

●

●

●●●●

●

●

●

●

●

●

●

●

●●

●

●

●

●●●

●●

●

●● ●●●

●● ●

●

● ●● ● ●

●

●●●●●

●

●

●● ●

●

●

●●

● ●

●

●

●●

●●●●

●●●●

●

●●●●

●

●●

●

●

●

● ●●

●

●

●

●

●●

●●●

●●

●

●

●●

●

●●

●●●●

●

●●●● ●

●●

●

●

●●

●● ●●

●

●

●●

●

●

●

●

●●

●

●

●●

●

●●

● ●●

●

●●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●

●

●

●

● ●

●●

●●

●

●●

●

●●

● ●

●●

●

●

●●●

●

●●

●

●

●●

●

●

●

●●

●

●

●●●

●●

●

●

●●● ●

●

●●

●●

● ●●

●

● ●●

●

●●

●

●

●●

●●

●

●

●●

● ●●●●

●●

●

●

●

●

● ●●●

●●

●●●

●●

●● ●●

●

●●●

●●

●

●

●●●

●

●●

●

●●

●

●●

●

●

●●● ●

●

●●

●

●

●

●

●●●

●●

●

●

●

●

●●●

●●

●●

●●●●

●●

●

●

●●●

●

●

●● ●●

●●●

●

●

●

● ●

●

●●

●●

●●

●

● ●

●●●●

●

●

●●

●●

●

●

●

●

●●●●

●●●

●

●

●

●●

●●

●●

● ●●

●●

●

●

●

●

●

●●●●

●

●●●

●

●

●

● ●

●

●

●

● ●●●

●●●

●

● ●●●●

●

●

●●

●●

●

●

●●

●

●●●●●

●

●

●●

●●

●●●

●

●

●

●● ●

● ●●●●

●●

●

●

●●●●

● ●

●

●●●●

●●●

●

●

●

●

●●● ●

●

●●

●●●●

●

●

● ●●

●

●

●

●

●

● ●●

●●● ●

●

●

●

●●

●

●

●

●

●

● ●

●

●

●●●●

●

●

●

●

●●

●●

●

●

●●● ●●●

●

●●

●

●

●●●

●

●●●

●

●

●

●

●

●

●

●●●●

●●

●

●

●

●●

●●

●●●

●●

●●●●

●

●

●

●

●

●

●

●

●●

●

●●

●

●

●

●

●

●●●

●●●

●

●●

●

●

●●●●

●

●

●

●

●●

●

●

●

●

●

●

●●

● ●●

● ●●●

● ●

●

●● ●●

●

●

●●

●

●

● ●

●● ●●

●

● ●

●

●● ●● ●●●

●

●

●● ●

●●

●

●

● ●

● ●●

●●

●● ● ●●

●

●●

●

●

●

●●

●

●

●

●

●

●●●

●

●

●●● ●●

●●●

● ●●

●

● ●●

●● ●●

●●●

●

●●

●●

●

●●

●

●

●●

●

●

●

●●

● ●

●

●●

●

● ●

●●

●

●

●

●●●●

●

●

●

●

●

●

●

●

● ●●

●●

●●

●

●●

●

●

●●● ●

●

●●●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

● ●

●

●● ● ●●●

●

●

●● ●

●

●

●●

●

●●

●●●

●● ●

●●

● ●

●

●

●

●

●

●

●●

●●●●

●●●

●

●

●

●●

●

●

●

●●● ●

●

●●

●●●

●● ●

●

●

●

●

●

●

●

●

●

● ●

●

●●

●

●

●

●●

●●

●●

●●

●●

●

●●

●●

●●

●

●●

●

●

●

●● ●● ●

●

●

● ●●

●

●●

●

●●

●

●

●

●

●●

●●

●●●

● ●

●

●

●

●

●●

●

●

●●

●●●

●

●

●●●●

●●

●

●

●

●

●

●

● ●● ●●

●

●

●

●

●●

●

●●

●

●

●

●

●

●

●●

●●

●●●

●

●

●●

●

●●

●●

●

●●

●

●

●

● ●●●

●●

●

●●

●●

●

● ●

●

●

●

●

●

●●

●

●

●● ● ●

●● ●

●●

●

●

●

●●●

●●

●

●

●

●●●

●

●

●

●

●●

●

●●● ●

●●●

●

●

●●

●

●

● ●●

●

●

●● ● ●

● ●●

●●

●

●●

●

●

● ● ●●

●●

●

●●

●

●

●

●

●

● ●

●

●

●

● ●

●

●

●

●●

●●

●●● ●●

●

●

●●

●

●

●

●

●●

●

●

●

●

●●●●

●

●

●

●● ●●

●

●

●

● ●

●

●

●●

●

●●

●

●

●●

●

●

●

●

●●●●●●

●

●

●

●●

●

●●

●

●

●●●

●

●● ●●

●

●

●

●●

●

●

●

●

●

●

●●

●●

●

●

●

● ●●●●

●

●

●

●

●●

●

●

●

●

●

●

●● ●●

●

●

●●

●●

●●●

●

●●●

●

●●●

●

●

●●

●

●●●●●●

●

●

●

●

●●

●●

●●

●

●

●

●

●

●●●

● ● ●

●

●

●

●●

●●●●●●

●●●

●

●●

●●●

●

●

●

●●

●

●●

●

●

●●

●●

●

●

●

●

●

●●

●

●

● ●

●

●

●

●●●

●● ●●

●

●

●●

●

●●

●

●

●

●

●

●●● ●

●

●

●

●

●

●

●●●

●●

●●

●

●

●

●

●

●

● ●

●●

●

● ●

● ●

●

●

●

● ●●

●

●●●

●●

● ●●

●●

●

● ●●●

●●● ●

●●

●●●

●

●●

●

●

●●●

●

●●

●●●

●

●

●●

●

●●

●● ●

●●●

●

● ●

●

●

●

●●

●●

●

●

●

●●●●

●●●

●

●●

●●

●

● ●●

●●

●●

●

●●

●

●●●

●● ●

●

●

●

●●

●

●

●

●● ●●

●

●●●●

●

●●

●

●

●

●

●

●●●●

● ●●

●

●

●

●

●

●●

●

●

●

●●●

●

●

●●

●

●

●

●

●

●

●●●●●

●

●●●

● ●

●

●

●

●●

●●

●

●

●●● ●●

●

●

●

●

●

● ●

●●

●

●

●

● ●●●

●

●

● ●

●

●●

●

●

●

●

●

●

● ●

●

●●

●

●●

●

●

●

●●

●●

●

●●●●

●

●

●

●

●● ● ●●

●

●●●

●●

●

●●●●

●

●●

●●●

●

●

●

●●●●

●●

●

●

●

● ●●

●

●●

●

●●

●●

●

●

●●

●

●

●

●●

●

●

●

●● ●

●

●● ●●

●●●

●●●

●● ●

● ●●

●

●

●

●● ●

●● ●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

● ●●

●

●●

●

●●

●

●

●

●

●

●

●

●●

●

●

● ●

●●

●●●

●

●

●

●

●●

● ●

●●

●

●●●

●

●

●●

●

●●●

● ●●●

●●●

●●●

●

●

●

●

●

●

●●

●●●

●

●●

●●

●

●

●

●

●

●

●●

●

●●●●

●

●

●

●

●

●

●●

●

● ●●●

●●●

●●

●●

● ●

●

●

●●●

●●

●

●

●

●

●

●●

●●

● ●●

●●●

●

●

●●

● ●●●

●

●●

●

●

●

●

●●

●●

●

●

●

●

●●

●

●

●

● ● ●●

● ●●●

●

●●

●

●

●

●

●

●

●

●

●●

●

●

●●

●

●●

●

●

●

●

●●

●● ●●●

●

●

●

●

●●

●

●

●

●

●

●

●●●

●

●

● ●●●

●

●

●●●

●●

● ●●

●●

●●●●●

●

●

●●●

●

●

●●

●

●

●

●●●

●●

●

●

●

●

●●

●

●

●●●

●

●

●

●

●

●●●

● ●●

●●

●

●

● ●●

● ●

●

●

●

●

●

●

●

● ●

●

●●●

●

●

●

●●●●

●●

●●●●

●●

●

●●●

●

●●

●●

●

●

●●

●

●

●●

●

●

●●

● ●●

●

●

●

●●●

●●

●

● ●

●●

● ●

●●

●

● ●

●

●●

●

●

●●

●

●

●

●

●

●

●●

●

●

●●

●●

● ●

●

●●

●●

●

●●

●

●

●

●●●

●

●

●●

●

●

●● ●

●

●

●

●

●

●●

●●●●●

●

●

●●

● ●●

●●

●●●

●●●

●

● ●●

●

●●

● ● ●

●

●

●●●

●●

●●

●● ●●

●

●

●

● ●●●

●

●

●

●● ●

●

●●

●

●

●●

● ●

●

●

●●

● ●●

●

● ●

●

●

●●

●●

●

●

●●

●

●

●

●●

● ●●

●

●● ●●●●

●

●

●

●

●

●●

●

●

●

●●

●

●

●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●

●

●

● ●●

●

●

●●●

● ●

●●

●

●●●

●

●●

●●

●

●● ●●●

●

●

●

●

●

●

●

●●●

●●

●

●●

● ●●

●

●

●●

●●

●●●

●●●

●●

●

●●

●

●●

●

●

●

●●

●

●

●

●

●

●●

●

●

●

●

●

●●

●●

●●●●

●

●

●●●

●

●●

●●

●

●●

●●

●

●

●

●

●

●

●

●

●●●●

●

●● ●●

●●● ●

●●●

●

●●

●

●

●

●●

●

●●

●●

●●

●

●●●

●

●

●

●●●

●

●● ●

●●●●

●●

●●

●

●

●

●● ●

●

●

●●

●

●

● ●

●

●

●

●● ●●●●

●● ●●

●

● ●●

●

●

●

●

●

●

●

●

●

●● ●●●

●

●

●

●

●●

●

●●

●

●

●

●

●●

●

●

●●

●●

●

●●

●●●

●

●

●

●

●

●●

●●

●●●

●

●●

●●●

●●

●

●●

●

●●●

●

●●

●

●

●

●

●●

●● ●●

●

●●●●

●●

●● ●●

● ●

●

●●

●

●●●

●●

●

●

● ●●●

●

●●

●

●

●●

●

●●

●

●

●

●

●

●

●

●●

● ●●●

●

●●

●

●

●●● ●

●

●●

●

●

●

● ●

●●●● ● ●

●

●● ● ●●

●●●

● ●

●

●

●

●●

●●

●

●●●

●

●●

●

●●●●

●●

●

● ●●

●

●

●

●●

●●

●

●●●

●●●●

●●

●

●●

●

●

●

● ●●● ●●●

●●

●●●●

●●●

●

●● ●●

●● ●

●●

●●●

●

●

●

●

●

●

●

●●

●

●

● ●●

●

●●

●

●

●●●

●●

●

●●

●●

● ●●

●●

●

●

●

●

●

●● ●●

●

●●

●

●

●●

●● ●

●●●

●●

●●●

●●

●

●

●

●●

●

●

●

●

●●●●

●

●

●

●●

●

●●

●●●

●

●● ●

●

●

●

●●

● ●●●

●●●●

●●●

●●●●

●

●●●

●● ●

●

●

● ●

●

●

●

● ●

●●●●

●●●●

●

●

●

●

●

●

● ●●

●

●

●

●

●

●

●

●

●

●●

●●

●

●●

● ●●

●●

●●

●

●

●

●

●

●●●

●●

● ●

●

●

●

●

●● ●●●●●●●●●●

●●

●

●

●

●

●

●

●●

●

●●

●

●●●●

●●

●

●

●

●

●

●●

●

●●

●●

●●●●

●

●

●

●●

●

●

●

●●

●●

●

●●

●

●

●

●

●

●

●

●●●

●

●

●

●

●

● ●

● ●●● ●

1 2 3 4 5

0.0

0.2

0.4

0.6

0.8

1.0

自社での累積購買回数(対数)

購買し

た商

品カ

テゴ

リの

多様

性(H

HI)

図 1 顧客行動の多様性と購買行動の関係

量的変数を組み合わせて算出されるより詳細な情報であ

るといえる．従って，場合によっては多様性変数を予測

に用いることで，このような量的変数がモデルにとって

のノイズになることも考えられる．従って，実施する分

析においては従来的な変数を除き，その代わりに多様性

変数を用いたモデルについても作成する．

4. 解析例 1: ECサイト上での実購買の予測

まず解析例 1では，ECサイト上でのWeb閲覧データから抽出した購買頻度データを用いて予測を行う．本解

析の目的，用いたデータと予測モデルについて示し，最

後にその結果を考察する．

4 ·1 解 析 の 目 的

本解析では国内の大手 ECサイトを対象として，サイト内で購買を行ったユーザを企業の自社顧客として扱い，

彼らの将来の購買についての予測を行う．これは多様性

変数が顧客の将来購買の予測にあたって有用であること

を示すことを目的としている．またその際には，既存研

究 [新美 15]と同様の分析を Deep Learningとパラメトリックな統計モデルのそれぞれで実施し，その予測精度

を比較する．また，マーケティングの研究分野において

は，消費者の行動における購買の量や次の訪問・購買ま

での時間・頻度といった量的な情報に関心を持って研究

が行われる場合が一般的であるため，本研究では購買有

無や訪問有無ではなく，これらを量的変数のまま予測を

行った．

4 ·2 使 用 デ ー タ

解析例 1で用いたデータは，株式会社ビデオリサーチインタラクティブが調査パネル対象者から収集したWebアクセスログである．このデータは実購買等のトランザ

クションデータは持たないものの，日本のインターネッ

ト利用の人口統計に基づき代表性を担保して抽出された

13000人程度のユーザが PCで閲覧したWebサイトについて，アクセスされた日時・URL単位での滞在時間（秒数）・リファラ情報などがサイト横断的に収録された大規

顧客行動の多様性変数を利用した購買行動の予測 5

模な User-Centric Clickstream Dataである．本解析においてはこのデータを特定のWebサイトに限定することにより，企業が得られる擬似的な Site-Centric ClickstreamDataとして扱う．実際の解析では今回使用したパネルデータに参加して

いる対象者のうち，1年以上の期間にわたって対象の ECサイトにアクセスしていること，直近の購買間隔を算出す

るために累積で 2回以上購買していることを満たす 4146人である．

購買の有無については，ECサイトで実際に購買した後に表示されるサンキューページの URLへのアクセスをもって購買と判断しており，誤操作や買い忘れを考慮

するため，1日に複数回発生した購買については 1回の購買と判定している．

4 ·3 使 用 モ デ ル

本研究では，Deep Neural Network (DNN)の代表的なモデルとして，Feed-Forward Neural Network (FFNN)による 3層の隠れ層と各層に 10個の素子 ∗5を持つ学習モデ

ルを作成した．学習方法にはベイズ正則化 (Bayesian Reg-ularization)を用いている．これは，Levenberg-Marquardt法 (More 78)∗6を元に，過学習を防ぐために損失関数に正則項を設定した手法である．また，BIC等の情報量基準を用いて汎化誤差の上昇前に学習を停止するため汎化性

能を高めることが可能であり，通常の誤差逆伝播モデル

より頑健であるとされる [Burden 09]．本モデルで用いた m個の入力 xi とそれに対応する重

み付け w ji，素子ごとのバイアス b jを持つある素子 h jの

出力は，

h j = tansig(b j +

m∑i=1

w jixi) (2)

と表せる．ただし tansigは今回Activation Function（活性化関数）∗7として用いている Hyperbolic tangent sigmoidfunctionであり，以下のように示される．

tansig(x) =2

1+ e−2x − 1 (3)

それぞれの解析では，2 つの目的変数「将来 1ヶ月間の購買回数」，「将来の訪問期間」についての予測を行う．

使用データから顧客ごとに最新 12ヶ月間の ClickstreamDataを取得し，その期間を最初の 11ヶ月 (t1)とそれ以降の最新 1ヶ月間 (t2)に分割した．顧客の将来の行動を予測するため，t1 を既知の期間，そして t2 を将来の未知の

∗5 ニューラルネットワーク内の構成要素であり，ニューラルネットワークはこの演算素子を層状にした階層構造が互いに結合した構成となっている [萩原 06]．∗6 最急降下法と Gauss-Newton法を組み合わせた，非線形型関数の最小化の標準的な手法．∗7 前層からの出力を重み付きの線形和として入力することで当該素子の出力を生成する関数．例としてステップ関数を用いると出力が 0/1の二値になるなど，活性化関数により様々な素子を表現することが可能．

期間としている．まず「将来 1ヶ月間の購買回数の予測」では，自社における顧客の将来の購買回数の予測を行う

ため，既知の学習データである 11ヶ月間の購買データやその他の行動データから学習を行った上で，将来 1ヶ月に何回の購買が行われたかを予測している．次に「将来

の訪問期間の予測」では，既知の期間に観測された最後

の訪問日時から数えて何日目に再度訪問したかを予測し

た．未知のデータとして扱っている将来 1ヶ月間の間に複数回の訪問があった場合には，既知期間から最も直近

の 1訪問のみを対象とした．本研究では顧客行動の予測における多様性変数の有効

性を評価するにあたり，投入する変数の組み合わせを変

えた複数の予測モデルを作成している．用意した「基本

変数」，「最頻値変数」，「多様性変数」の 3つの分類から，基本変数に加えて最頻値変数と多様性変数をそれぞれ投

入したモデルと両方投入したモデルの計 4モデルを作成した．訪問時期と累積購買回数ともに予測値と実測値で

の RMSE (Root Mean Squared Error)を算出しモデル間で比較した際に，多様性変数が説明変数として用いられ

ているモデルで最も RMSEが改善していることをもって多様性変数が有効であると判断する．

また，比較に用いるパラメトリックな統計モデルでの

予測には，GLM (Generalized Linear Regression)による回帰分析モデルを用い，最尤推定による推定値を利用し

た．解析では目的変数として「累積購買回数」「将来の購

買期間」を採用しているが，それぞれの予測モデルにつ

いて一般的に利用されるモデルとしてポアソン分布によ

る回帰モデルとワイブル分布による回帰モデルを用いた．

予測精度の算出の際には，DNN モデルとの公平性を期すため，元データのうち 75 %のデータからモデリングし，作成したモデルに対して残り 25 %のテストデータを適用することで予測を行い，RMSEを算出した．

4 ·4 使用した変数

ここで本解析で用いた変数の一覧を表 2に示す（但し解析例 1と解析例 2で用いた変数をともに表示している）．変数は大きく 3つに分類されており，それぞれ「基本変数」，「最頻値変数」，「多様性変数」としている．まず基

本変数には，過去の購買・Webサイトへのアクセス・モバイルアプリの使用の頻度情報などを始めとして，基本

的な顧客情報を表しうる量的変数が主となっている．次

に多様性変数には，どれほど幅広い商品カテゴリを購買

したか，どれほど特定のカテゴリに依存しているか，と

いった質的な情報を多様性変数として集めている．また

最頻値変数では，最もよく購買された商品カテゴリや最

もよく来店した自社店舗など，顧客の嗜好を表す変数が

中心となっている．最頻値変数は全て Deep Learningモデルに投入しやすいようダミー変数としている．本来で

あればこの最頻値変数の中に「t1 期間に自社で最も購買された商品カテゴリ」や「t1 期間に最も来店した自社店

6 人工知能学会論文誌 32巻 2号 B（2017年）

表 2 用意した変数一覧

変数名 変数使用 説明

解析 1 解析 2

目的変数

Buyt2 ◯ ◯ 目的変数 1: t2 期間の自社での累積購買回数Durationt2 ◯ ◯ 目的変数 2: t2 期間中の最新の訪問間隔（日数・対数化）

基本変数

Buyt1 ◯ ◯ t1 期間の自社での累積購買回数Durationt1 ◯ ◯ t1 期間の自社での平均購買間隔（日数・対数化）Accesst1 ◯ ◯ t1 期間中に自社へアクセスした平均間隔（日数・対数化）PriceDi f f erencet1 - ◯ t1 期間中に購買された商品価格の最大値と最小値の差BuyQuantityt1 - ◯ t1 期間中の 1取引での平均購買品目数CheckinRanget1 - ◯ t1 期間中のチェックイン店舗の範囲AppBrowseProductt1 - ◯ t1 期間中にモバイルアプリで閲覧した商品数Actiont1 t1 期間中の次のそれぞれの行動の回数

◯ - - 1: サイト内での商品検索◯ ◯ - 2: 自社Webサイトへの流入- ◯ - 3:モバイルアプリの起動- ◯ - 4: 店舗への訪問- ◯ - 5: アプリでの商品詳細の閲覧

最頻値変数

MostRe ft1Web ◯ - t1 期間に最も使用した自社サイトへの流入経路MostT imet1Web ◯ ◯ t1 期間に自社Webサイトへ最もアクセスした時間帯MostWeekdayt1Web ◯ ◯ t1 期間に自社Webサイトへ最もアクセスした曜日MostT imet1App - ◯ t1 期間に自社アプリを最も使用した時間帯MostWeekdayt1App - ◯ t1 期間に自社アプリを最も使用した曜日MostT imet1Visit - ◯ t1 期間に自社店舗へ最も来店した時間帯MostWeekdayt1Visit - ◯ t1 期間に自社店舗へ最も来店した曜日

多様性変数（HHI／累積パーセント）Re f erert1 ◯ - t1 期間の自社Webサイトへの流入経路の多様性AccessT imet1Web ◯ ◯ t1 期間の自社Webサイトへアクセスした時間帯の多様性AccessWeekdayt1Web ◯ ◯ t1 期間の自社Webサイトへアクセスした曜日の多様性ViewCategoryt1 ◯ ◯ t1 期間に自社で閲覧した商品カテゴリの多様性PurchaseCategoryt1 - ◯ t1 期間に自社で購買された商品カテゴリの多様性VisitS toret1 - ◯ t1 期間に来店した自社店舗の多様性AccessT imet1App - ◯ t1 期間の自社アプリを使用した時間帯の多様性AccessWeekdayt1App - ◯ t1 期間の自社アプリを使用した曜日の多様性AccessT imet1Visit - ◯ t1 期間の自社店舗へ来店した時間帯の多様性AccessWeekdayt1Visit - ◯ t1 期間の自社店舗へ来店した曜日の多様性

舗」についても投入すべきであったが，本研究で対象に

している企業の展開する店舗数が非常に多く，次元数が

極めて増加するため今回は分析から除いた．

一部に解析例 1と解析例 2の一方でしか用いていない変数があるが，これらは各解析例で異なるデータを使用

していることから，データ上の都合として測定が不可能

であったことを示す．例として純粋な Clickstream Dataのみで解析を行っている解析例 1では価格に関する情報は得られず，解析例 2では Clickstream Dataの都合からリファラ情報を得ることができなかった．

4 ·5 解 析 結 果

将来の購買回数と訪問期間について，DNNとGLMを用いて予測した結果をそれぞれ表 3，表 4に示す．まずRMSEの値が最も改善したのは，DNNのモデルを用いた際に基本変数に加えて多様性変数のみを説明変数とし

て加えたものであった．解析結果からわかる特徴につい

て述べると，DNNと GLM間での予測結果の比較では，いずれの場合においても常に DNNの方が予測精度が改善している．また DNNについては購買回数の予測に比べて訪問期間を扱った場合には予測精度の改善幅が小さ

くなっている．これらの特徴については解析例 2での結果も加味して 6 ·2節にて考察する．

表 3 解析例 1-1: 将来の購買回数予測精度の比較（RMSE）

予測方法 基本 基本+最頻 基本+多様性 基本+最頻+多様性

DNN 1.57 1.57 1.38 1.84GLM 2.36 3.07 2.42 3.12

表 4 解析例 1-2: 将来の訪問期間予測精度の比較（RMSE）

予測方法 基本 基本+最頻 基本+多様性 基本+最頻+多様性

DNN 1.16 1.17 1.14 1.17GLM 2.19 2.19 2.19 2.19

5. 解析例 2: 小売企業の実データを用いた予測

次に解析例 2では，プライベートブランドを扱う小売企業の収集した複数のデータを用いて同様の分析を行う．

5 ·1 解 析 の 目 的

顧客のアクセスログや ECサイトでのトランザクションデータなどのWeb情報は，既存研究や解析例 1をはじめとして一般的には消費者のオンライン上での行動の予

測に活用されることが多い．しかし，Web上での顧客の行動を継続的に取得した場合，アクセスしやすい時間帯

やよく閲覧している商品カテゴリ等，顧客の行動や趣向

に関して店頭の POSデータだけでは得ることの難しい多様な情報を得られるであろう．これは必ずしもオンライ

ンの行動予測のみに有用であるとは限らず，むしろ実店

舗での顧客行動においても広く有用である可能性が考え

られる．そこで本解析では，アクセス履歴等のWebデータから顧客ごとの行動パターンを発見し分析に考慮する

ことが，Web以外の実店舗での購買や訪問間隔等の行動を予測するにあたっても広く有用であることを示す．ま

た解析例 1と同様に，参考として GLMを用いた予測も行うことで結果を比較している．

5 ·2 使 用 デ ー タ

本解析では，自社プライベートブランド商品を販売す

る小売企業から提供された各種データを使用した．この

データには顧客のデモグラフィック情報（年齢，性別，居

住都道府県等），自社 ECストアと自社店舗でのトランザクションデータ（受注 ID，受注日時，販売商品の JANコード，商品ジャンル，価格，割引の有無，返品の有無

等），アプリ内での行動履歴（アプリ起動，チェックイン

店舗コード，閲覧商品等），自社Webページ内の閲覧履歴（閲覧URL，欲しいなどのコメント等）の情報が含まれている．そしてこれらデータは，顧客がユーザ登録を

行っている場合には，全て共通のユーザ IDによって紐付いており横断的に参照することができる．従って，多

くの顧客についてWebでの商品情報の閲覧や，購買の有無に関わらない実店舗への訪問，さらには最終的な購買

の有無など，オンライン・オフラインを問わず行動を時

系列に従って継続的に追跡することが可能となっている．

顧客行動の多様性変数を利用した購買行動の予測 7

購買を伴わない店舗への訪問は，モバイルアプリによる

店舗への「チェックイン機能」から判別が可能である．顧

客はチェックインごとに特典ポイントを得られることか

ら，顧客は来店した際に積極的にチェックインを行うイ

ンセンティブを持つ．

実際の解析では，この使用データからいくつかの条件

に従ってユーザを抽出した．まず，分析対象者がユーザ

登録を行っていること．これは将来的な訪問や購買の有

無／回数を予測するにあたって，ユーザ IDの登録により継続的に購買情報が得られている必要があるためであ

る．また，多面的な分析を行うため，Webアクセスやスマートフォンアプリとの連携が取れている顧客に絞って

分析を行う．さらに，将来の行動の予測を行うにあたって

十分に顧客行動の分析が行えるよう，Web閲覧やアプリの使用を含めて少なくとも 1年間以上継続的にデータを得られた顧客のみを扱う．また，得られたWeb閲覧データは，自社サイトに限定された Site-Centric ClickstreamDataである．調査期間が 1年以上でありながらアクセス回数が極端に少ない顧客については分析が困難であると

判断し，対象から除外した．結果として分析対象の顧客

は 56710人となった．使用した変数は表 2 に示しているが，設定した説明変数としてCheckinRanget1 では顧客がチェックインした

全店舗の地理的な幅広さを計測している．例として顧客

customerkが訪問した nk個の店舗 siの緯度，経度がそれ

ぞれ yi，xi の時，顧客ごとの訪問店舗の重心 Gk = (x, y)であり，重心と各店舗の距離 lengthi は

lengthi =

√(x− xi)2 + (y− yi)2 (4)

である．CheckinRanget1 では顧客ごとにこの値の最大値

を設定している．顧客の行動範囲が幅広いほど重心から

の店舗の距離が遠くなるため，CheckinRanget1 で顧客の

地理的な行動範囲の大きさを数値化できると考えられる．

5 ·3 解 析 結 果

表 5，表 6に，解析例 2における「将来 1ヶ月間の購買回数の予測」，「将来の訪問期間の予測」それぞれの予

測精度の一覧を示した．解析例 1の結果と同様に，DNNを用いて基本変数に加えて多様性変数のみを加えたモデ

ルで RMSE が最も改善する結果となった．解析例 1 と同様の点としては，DNN について購買回数の予測に比べて訪問期間を扱った場合に予測精度の改善幅が小さく

なっている点がある．一方で解析例 1と異なる点として，GLMで多様性変数を用いた場合に購買・訪問ともに予測精度が改善している．

6. ま と め

本章では 2つの解析例の結果から，非線形の予測モデルとしての DNNと多様性変数の有用性について考察する．

表 5 解析例 2-1: 将来の購買回数予測精度の比較（RMSE）

予測方法 基本 基本+最頻 基本+多様性 基本+最頻+多様性

DNN 3.10 3.78 2.84 4.54GLM 4.19 4.20 4.10 4.32

表 6 解析例 2-2: 将来の訪問期間予測精度の比較（RMSE）

予測方法 基本 基本+最頻 基本+多様性 基本+最頻+多様性

DNN 1.03 1.08 1.01 1.05GLM 4.21 4.23 4.09 4.86

6 ·1 多様性変数の比較検討

本研究では顧客の行動の幅広さを「多様性」として定

義して分析に考慮したが，このような顧客行動の多様性

を測定する指標としてHHIを用いることを提案した．これは要素ごとの不平等を計測できる指標として適切であ

ると考えたためである．しかしながら，同様の目的で使

用できる指標には，HHIの他にも所得の不平等を計測し数値化するジニ係数 [Weymark 81]も考えられる．そこで，ここまで扱ってきた解析例についてHHIとジニ係数で比較検討するため，表 7にそれぞれの指標を説明変数として用いた場合の RMSEを示した．いずれの解析についても，HHIの方がジニ係数よりも RMSEが改善しており，本解析における多様性変数の測定には，HHIがより適した指標であることがわかる．

6 ·2 多様性変数の有用性

まず得られた各解析例の結果から，解析に使用するデー

タ，モデル，解析内容による予測精度の変化については，

「特徴 1: すべての解析例において常に DNNで基本変数と多様性変数を用いた場合に最も RMSEが改善した」，「特徴 2: 解析例 1,2ともに DNNでは購買回数に比べて訪問間隔の予測時に多様性変数の改善幅が小さい」の 2点にまとめることができるだろう．

まず特徴 1に関してであるが，すべての解析例についてDNNとGLMのモデル間での予測精度の変化と，各同一モデル内で変数を変えた際の予測精度の変化について俯

瞰すると，全体的に DNNを用いた場合に RMSEは大きく低下しており，その中でも基本変数と多様性のみを用

いた場合が常に最良となっている．この結果から，DNNにおいて捉えられている現象がより単純なモデルとして

の GLMでは捉えきれていないといえる．また GLMでは解析例 1-1など一部の結果において基本変数と多様性を用いたモデルよりもその他のモデルの方が予測精度が

向上している場合があるが，いずれにしても DNNでの場合と比べて大幅に予測精度は悪く，あくまで現象を説

明できていないモデル内での挙動である．

次に特徴 2に関しては，企業にとっては訪問間隔よりも購買量がより重要であるが，購買と訪問という行動自

体の差異に着目した場合，Webサイトや実店舗といった形態にかかわらず，購買は単なる訪問と比較すると意思

8 人工知能学会論文誌 32巻 2号 B（2017年）

表 7 DNN モデルにおけるジニ係数と HHI での RMSE の変化

使用変数 解析 1-1 解析 1-2 解析 2-1 解析 2-2

基本変数 + HHI 1.38 1.14 2.84 1.01基本変数 +ジニ係数 1.57 1.18 2.92 2.50

決定と費用負担が発生するという点がある．店舗やWebサイトへの訪問では自社以外を含めて複数のサイトや店

舗に様々に訪問しえるが，商品購買についてはある商品

カテゴリについて特定の店舗で購買した場合に他の店舗

でも購買するということは少ない．DNNの結果からは，コストを伴わない訪問の間隔は最頻値変数という単なる

個人のWeb等での活動量の大きさで説明できるが，購買回数という意思決定と費用を伴う行為が多様性変数を用

いることでより予測精度が向上するということは注目に

値する．これは今後さらに研究する意義があるが，多様

性変数は (より多様な情報源から探索しようとするなど)自社以外での行動要素が含まれているということを示唆

している．

さてここまでで各解析結果から多様性変数を用いたモ

デルの挙動について考察を行った．結果として将来の行

動を予測するにあたって既に得られている顧客行動から

多様性変数を算出することで予測精度の向上をはかるこ

とができた．一般的には顧客の行動についての情報を得

るためには，アンケートによる情報収集をはじめとして

追加的なコストや顧客への負担のかかるものも多い．そ

ういった中で本研究ではモバイルアプリやオンラインで

の閲覧履歴，実購買のデータなど，いずれも企業が通常の

企業活動を行う上で得られるであろう顧客行動の実デー

タのみから計測した多様性変数を解析に用いている．そ

のような点からも，本研究で提案する多様性変数は企業

の実務においても十分に活用できるものであると考えら

れる．

6 ·3 DNNを用いた予測モデルの構築について本研究では DNNでの予測を行うにあたって多様性変

数を用いることに加えて，線形モデルとしての GLMとの予測精度の比較を行った．解析ではいずれも基本変数に

加えて最頻値変数を用いた場合には常に予測精度が悪化

している．これは従来マーケティングを行うにあたって有

用であると考えられてきたような変数が，Deep Learning等のモデルにおいてはむしろノイズになってしまう可能

性を示唆している．ひとつの考えられる原因として「最

もよく利用する店舗」や「最もよく購買する商品カテゴ

リ」などをダミー変数として投入する場合，展開してい

る店舗数や取り扱う商品カテゴリの数によっては入力の

次元数が非常に大きな値になってしまうことが挙げられ

るだろう．最頻値変数等を用いずとも DNNで多様性変数を用いることで，より少ない変数で柔軟に予測が可能

な点からもビッグデータ時代の莫大なデータ量に対する

効率的な予測モデルの構築が可能と考えられる．

表 8 解析例 2-1での隠れ層の数と予測精度，解析所要時間の関係

隠れ層 1層 3層 5層 10層

RSME 17.35 2.84 3.41 2.85所要時間 3分 8分 15分 40分

また，本研究における DNNのモデルでは隠れ層を 3層持つ FFNNを用いているが，本研究の実施にあたって予め網羅的に解析を行っている．参考までに表 8に隠れ層の数と学習の所要時間，そして予測精度について示す．

隠れ層を増やした場合には解析時間が莫大に増加する一

方で予測精度には大きな変化が見られず，本研究では 3層の FFNNを用いることが最適であると判断した．

6 ·4 限 界 と 今 後

今回は使用しているデータ上の都合から，解析例 2の小売店での購買データについては，Clickstream Dataの分析にあたって通常であれば得られるはずのリファラ情

報を分析に考慮することができなかった．ECサイトにおいてリファラ情報は顧客の自社外での行動に関して企業

が得られる数少ない情報である．実際に企業で行われる

分析ではサーバから得ることが可能であるため，リファ

ラの多様性を分析に加えることで更に予測精度が向上す

る可能性がある．

また，今回分析対象とした企業ではモバイルアプリへ

のプロモーション情報の通知なども行っている．さらに

Web上では季節に応じたキャンペーンや特集ページなど，様々な企画が行われている．本研究では分析モデル上の

限界からこのような情報については加味しなかった．し

かしながらこのようなマーケティング情報の受信により

来店までの期間や購買頻度が変化する可能性も考えられ

るため，今後はさらに幅広い情報を用いて分析を行う必

要がある．またマーケティング情報としては，閲覧した

商品の詳細情報から word2vecを用いて商品の情報を特徴量ベクトルとして加味することで，カテゴリよりもさ

らに詳細な情報を分析に取り入れることも可能であると

考えられる．

謝 辞

本論文の執筆にあたって編集担当委員および 2名の匿名の査読者の方より有益なコメントをいただきました．

また株式会社ビデオリサーチインタラクティブより貴重

なデータをご提供いただきました．この場を借りて御礼

申し上げます．

本研究は JSPS 科研費 JP26285151 及びリーディング大学院プログラム PhDプロフェッショナル登龍門（名古屋大学）の助成を受けたものです．

顧客行動の多様性変数を利用した購買行動の予測 9

♢ 参 考 文 献 ♢

[Bengio 07] Bengio, Y., Lamblin, P., Popovici, D., Larochelle, H.,et al.: Greedy layer-wise training of deep networks, Advances in Neu-ral Information Processing Systems, Vol. 19, p. 153 (2007)

[Blattberg 08] Blattberg, R. C., Kim, B. D., and Scott, A. N.:Database Marketing: Analyzing and Managing Customers, Springer(2008)

[Boser 92] Boser, B. E., Guyon, I. M., and Vapnik, V. N.: A trainingalgorithm for optimal margin classifiers, in Proceedings of the 5thAnnual Workshop on Computational Learning Theory, pp. 144–152,ACM (1992)

[Bucklin 02] Bucklin, R. E., Lattin, J. M., Ansari, A., Gupta, S.,Bell, D., Coupey, E., Little, J. D., Mela, C., Montgomery, A., andSteckel, J.: Choice and the Internet: From clickstream to researchstream, Marketing Letters, Vol. 13, No. 3, pp. 245–258 (2002)

[Burden 09] Burden, F. and Winkler, D.: Bayesian regularization ofneural networks, Artificial Neural Networks: Methods and Applica-tions, pp. 23–42 (2009)

[Davis 09] Davis, P. and Garces, E.: Quantitative Techniques forCompetition and Antitrust Analysis, Princeton University Press(2009)

[Dembczynski 08] Dembczynski, K., Kotlowski, W., and Weiss, D.:Predicting ADS click-through rate with decision rules, in Workshopon Targeting and Ranking in Online Advertising, Vol. 2008 (2008)

[Goldberg 92] Goldberg, D., Nichols, D., Oki, B. M., and Terry, D.:Using collaborative filtering to weave an information tapestry, Com-mun. ACM, Vol. 35, No. 12, pp. 61–70 (1992)

[萩原 06] 萩原克幸：ニューラルネットワークの基礎と理論的に重要な課題 (ニューラルネットワークの応用と今後の発展),プラズマ・核融合学会誌, Vol. 82, No. 5, pp. 282–286 (2006)

[石垣 11] 石垣司,竹中毅,本村陽一：日常購買行動に関する大規模データの融合による顧客行動予測システム実サービス支援のためのカテゴリマイニング技術,人工知能学会論文誌, Vol. 26,No. 6, pp. 670–681 (2011)

[伊藤 16] 伊藤浩二,西土祥子,山崎智章：NTTコミュニケーションズにおける AIを活用した事業変革への挑戦 (特集 NTTグループにおける AIの取り組み), NTT技術ジャーナル, Vol. 28, No. 2,pp. 26–29 (2016)

[Kim 10] Kim, J. B., Albuquerque, P., and Bronnenberg, B. J.: Onlinedemand under limited consumer search, Marketing Science, Vol. 29,No. 6, pp. 1001–1023 (2010)

[LeCun 89] LeCun, Y., Boser, B., Denker, J. S., Henderson, D.,Howard, R. E., Hubbard, W., and Jackel, L. D.: Backpropagationapplied to handwritten zip code recognition, Neural Computation,Vol. 1, No. 4, pp. 541–551 (1989)

[松村 16] 松村直樹,和泉潔,山田健太：POSデータに基づく欠品時の顧客行動を考慮した小売店舗の購買シミュレーション,人工知能学会論文誌, Vol. 31, No. 2, pp. 1–8 (2016)

[Mikolov 13] Mikolov, T., Sutskever, I., Chen, K., Corrado, G. S., andDean, J.: Distributed representations of words and phrases and theircompositionality, in Advances in Neural Information Processing Sys-tems, pp. 3111–3119 (2013)

[Moe 03] Moe, W. W.: Buying, searching, or browsing: Differentiat-ing between online shoppers using in-store navigational clickstream,Journal of Consumer Psychology, Vol. 13, No. 1, pp. 29–39 (2003)

[Moe 04] Moe, W. W. and Fader, P. S.: Dynamic conversion behaviorat e-commerce sites, Management Science, Vol. 50, No. 3, pp. 326–335 (2004)

[More 78] More, J. J.: The Levenberg-Marquardt algorithm: Imple-mentation and theory, in Numerical Analysis, pp. 105–116, Springer(1978)

[新美 15] 新美潤一郎,星野崇宏：ユーザ別アクセス・パターン情報の多様性を用いた顧客行動の予測とモデリング, 応用統計学, Vol. 44, No. 3 (2015)

[Park 12] Park, D. H., Kim, H. K., Choi, I. Y., and Kim, J. K.: Aliterature review and classification of recommender systems research,Expert Systems with Applications, Vol. 39, No. 11, pp. 10059–10072(2012)

[Poel 05] Poel, Van den D. and Buckinx, W.: Predicting online-purchasing behaviour, European Journal of Operational Research,

Vol. 166, No. 2, pp. 557–575 (2005)[Sharma 11] Sharma, A., Panigrahi, D., and Kumar, P.: A neural

network based approach for predicting customer churn in cellularnetwork services, International Journal of Computer Applications,Vol. 27, No. 11, pp. 26–31 (2011)

[Vieira 15] Vieira, A.: Predicting online user behaviour using deeplearning algorithms, CoRR, Vol. abs/1511.06247, pp. 1–12 (2015)

[Weymark 81] Weymark, J. A.: Generalized Gini inequality indices,Mathematical Social Sciences, Vol. 1, No. 4, pp. 409–430 (1981)

[Zhao 05] Zhao, Y., Li, B., Li, X., Liu, W., and Ren, S.: Customerchurn prediction using improved one-class support vector machine,in Advanced Data Mining and Applications, pp. 300–306, Springer(2005)

〔担当委員：馬場 雪乃〕

2016年 6月 23日 受理

著 者 紹 介

新美 潤一郎（学生会員）

2013 年名古屋大学経済学部経済学科卒業．2015 年同大学院経済学研究科産業経営システム専攻修士課程修了．同年より同専攻博士後期課程在籍．

星野 崇宏（正会員）

2004 年東京大学大学院総合文化研究科広域科学専攻博士課程修了．博士（学術）. 2004 年情報・システム研究機構統計数理研究所　助手. 2008 年名古屋大学 大学院経済学研究科准教授, 同年 JST さきがけ「知の創生と情報社会」領域 研究代表者. 2015 年慶應義塾大学経済学部教授．現在に至る. 社会科学における統計的因果推論,マーケティングサイエンス,行動経済学の研究に従事. 行動経済学会常任理事．2016 年日本学術振興会賞受賞