unsupervised evolutionary clustering algorithm for mixed type data

18
Intelligent Database Systems Lab 國國國國國國國國 National Yunlin University of Science and Technology 1 Unsupervised Evolutionary Clustering Algorithm for Mixed Type Data Zhi Zheng , Maoguo Gong , Jingjing Ma , Licheng Jiao , Qiaodi Wu 2010,CEC Presented by Chien-Hao Kung 2011/12/1

Upload: hamish-walls

Post on 02-Jan-2016

22 views

Category:

Documents


0 download

DESCRIPTION

Unsupervised Evolutionary Clustering Algorithm for Mixed Type Data. Zhi Zheng , Maoguo Gong , Jingjing Ma , Licheng Jiao , Qiaodi Wu 2010,CEC Presented by Chien-Hao Kung 2011/12/1. Outlines. Motivation Objectives Methodology Experiments Conclusions Comments. Motivation. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

國立雲林科技大學National Yunlin University of Science and Technology

1

Unsupervised Evolutionary Clustering Algorithm for Mixed Type Data

Zhi Zheng , Maoguo Gong , Jingjing Ma , Licheng Jiao , Qiaodi Wu2010,CEC

Presented by Chien-Hao Kung2011/12/1

Page 2: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.

2

Outlines· Motivation· Objectives· Methodology· Experiments· Conclusions· Comments

Page 3: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.

3

Motivation

· As a partitional clustering algorithm, K-prototype (KP) algorithm is a well-known one for mixed type data.

· However, it is sensitive to initialization and converges to local optimum easily.

Page 4: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.

4

Objectives

· In this study, KP is applied as a local search strategy, and runs under the Global searching to help KP overcome its flaws.

Page 5: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.Methodology

· K-prototype Algorithm─ Step1.Initializing.─ Step2.For each data item, calculating the distances.

─ Step3.Retest every data item.─ Step4.Repeat Step3. until no item changes its cluster.

5

Page 6: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.Methodology

· Evolutionary k-prototype(EKP)─ Step1 Initialization.─ Step2 Crossover.─ Step3 Mutation.─ Step4 KP Search.─ Step5 Evaluation and Selection.─ Step6 Termination Test.

6

Page 7: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.Methodology· Initialization

─ There are 8 parameters have to be set before evolution. Cluster number r is a weight in EKP which balance the influence on clustering Population size Proportion of initial individuals that generated by choosing

items randomly in dataset (IP) Crossover probability Mutation probability in simulated binary crossover(SBX) n in polynomial mutation

7

Page 8: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.Methodology

8

· Initialization─ Two kinds of random initialization schemes

─ The first is randomly choosing K data item as the prototypes of clusters

─ The second is randomly generating K prototypes

Ex:

[2.23,5.63],[6.56,5.13], and {1,2,3,4,5,6},{2,4}

=>{3.21,6.23,2,4}

Page 9: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.Methodology

9

· Crossover.─ Numerical type --Simulated binary crossover(SBX)─ Categorical type – Single point crossover

Page 10: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.Methodology

10

· Mutation

Page 11: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.Methodology

11

· KP Search· Evaluation and Selection

· Termination Test

Page 12: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.

· Parameter setting

Experiments

12

Page 13: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.Experiments

13

Page 14: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.Experiments

14

Page 15: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.Experiments

15

· Dataset

Page 16: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.Experiments

16

Page 17: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.

17

Conclusions· This paper propose a novel unsupervised clustering

algorithm for mixed type data named evolutionary k-prototype(EKP) .

· The experiment result show that the evolutionary framework improves the original algorithms markedly.

· EKP which can adjust this weight automatically needs to be studied.

Page 18: Unsupervised Evolutionary Clustering Algorithm  for Mixed Type Data

Intelligent Database Systems Lab

N.Y.U.S.T.

I. M.

18

Comments

· Drawback─ This method use the parameter too much.

Application─ Clustering