KNOWLEDGE BASED TECHNIQUESKNOWLEDGE BASED TECHNIQUES

23/4/2023/4/20 22

1. INTRODUCTION

many geographical problems are ill-structured

an ill-structured problem "lacks a solution

algorithm and often even clear goal achievement

criteria"

a DSS is one response to ill-structured problems

knowledge based techniques are another

23/4/2023/4/20 33

1.1 Example

where to put a label in a polygon? (the where to put a label in a polygon? (the ““label placement problemlabel placement problem””) - important in ) - important in designing map output from GIS designing map output from GIS

one rule might be "draw the label one rule might be "draw the label horizontally, centered at the centroid " horizontally, centered at the centroid "

there have been many attempts to there have been many attempts to reduce the label placement problem to a reduce the label placement problem to a set of simple rules and build these into an set of simple rules and build these into an "expert system" "expert system"

23/4/2023/4/20 44

1.2 Elements of knowledge based systems

techniques for acquiring knowledge

ways of representing knowledge internally

search procedures for working with the internally

stored knowledge

inference mechanisms for deducing solutions to

problems from stored knowledge

23/4/2023/4/20 55

1.3. Expert system “shells”

are software packages with functions which help are software packages with functions which help

the user construct special-purpose expert systems the user construct special-purpose expert systems

example applications of shells: example applications of shells:

1.building a system to make medical1.building a system to make medical

diagnoses - emulating the medical expert diagnoses - emulating the medical expert

2.building a system to emulate the cartographer's2.building a system to emulate the cartographer's

knowledge of map projections, to pick theknowledge of map projections, to pick the

best projection for a particular problembest projection for a particular problem

23/4/2023/4/20 66

2. KNOWLEDGE ACQUSIITION how is a knowledge base constructed? how is a knowledge base constructed? two approaches: two approaches:

by asking experts to break their knowledgeby asking experts to break their knowledge

down into its individual facts, rules etc. down into its individual facts, rules etc. by deducing rules from the behavior ofby deducing rules from the behavior of

experts experts by asking experts to break their knowledgeby asking experts to break their knowledge

down into its individual facts, rules etc. down into its individual facts, rules etc.

23/4/2023/4/20 77

2.1 Example of knowledge base constructed by experts

local government agency responsible for local government agency responsible for

regulating land use in vast sparsely populated regulating land use in vast sparsely populated

area - small staff area - small staff

decisions are subject to complex system of decisions are subject to complex system of

regulations, laws, past precedents, guidelines regulations, laws, past precedents, guidelines

decisions must be defensible in court decisions must be defensible in court

basic data - vegetation, soils, wildlife, geology basic data - vegetation, soils, wildlife, geology

etc. etc.

knowledge base of all regulations, laws, knowledge base of all regulations, laws,

precedents, guidelines precedents, guidelines

decisions can be generated from knowledge decisions can be generated from knowledge

base base

23/4/2023/4/20 88

2.2 Examples of knowledge inferred from interaction with

experts Knowledge Based GIS (KBGIS) developed Knowledge Based GIS (KBGIS) developed

by Smith and others by Smith and others system can reduce query time by system can reduce query time by

anticipating queries anticipating queries KBGIS analyzes queries received to KBGIS analyzes queries received to

"learn" about the pattern of queries and "learn" about the pattern of queries and organize its database to optimize response organize its database to optimize response

systems such as KBGIS learn about systems such as KBGIS learn about important spatial facts through the user's important spatial facts through the user's interaction with the system interaction with the system

23/4/2023/4/20 99

3. KNOWLEDGE REPRESENTATIONS

data structures in which knowledge data structures in which knowledge

can be stored can be stored

more general than conventional more general than conventional

databases databases

four general methods for representing four general methods for representing

knowledge - trees, semantic knowledge - trees, semantic

networks, frames, production rules networks, frames, production rules

23/4/2023/4/20 1010

3.1 Trees

way of organizing objects that way of organizing objects that are related in a hierarchical are related in a hierarchical fashion fashion

tree structures are common in tree structures are common in geographical data geographical data e.g. quadtrees and octrees e.g. quadtrees and octrees e.g. hierarchical nesting of census e.g. hierarchical nesting of census

reporting zones reporting zones

23/4/2023/4/20 1111

3.2 Semantic networks

knowledge is organized as a set knowledge is organized as a set ofof

nodes connected by labeled nodes connected by labeled links links

the GIS operations required to the GIS operations required to buildbuild

an information product from an information product from input input

data layers can be visualized data layers can be visualized as aas a

network of nodes and linksnetwork of nodes and links

23/4/2023/4/20 1212

3.3 Frames

usually consist of the name of a usually consist of the name of a phenomenon and the attributes phenomenon and the attributes that describe it that describe it

increasing availability of frame increasing availability of frame based expert system shells based expert system shells

23/4/2023/4/20 1313

3.4 Production rules

consist of two parts - situation part consist of two parts - situation part and action part and action part

most popular knowledge most popular knowledge representation in geographical representation in geographical applications applications

of the four areas of GIS - input, output, of the four areas of GIS - input, output, analysis and storage - output is most analysis and storage - output is most fully explored fully explored

23/4/2023/4/20 1414

4. SEARCH MECHANISMS

need a procedure for need a procedure for accessing knowledgeaccessing knowledge

each knowledge each knowledge representation has associated representation has associated search mechanismssearch mechanisms

23/4/2023/4/20 1515

5. INFERENCE is the creation of new knowledge is the creation of new knowledge

the solution to any problem is new knowledge which the solution to any problem is new knowledge which can be stored in the systemcan be stored in the system

a knowledge base can continue to grow as more a knowledge base can continue to grow as more knowledge is inferred from the existing baseknowledge is inferred from the existing base

e.g. a GIS can create new knowledge by computing e.g. a GIS can create new knowledge by computing topological relationships between objects from their topological relationships between objects from their geometrical relationshipsgeometrical relationships

deductive inference: deductive inference: creates new knowledge from existing facts through creates new knowledge from existing facts through

logical implication, e.g. using production ruleslogical implication, e.g. using production rules e.g. if A=B and B=C, then the system can deduce e.g. if A=B and B=C, then the system can deduce

that A=Cthat A=C inductive inference: inductive inference:

produces new generalizations ("laws") which are produces new generalizations ("laws") which are consistent with existing facts consistent with existing facts

e.g. if the database contains the knowledge that e.g. if the database contains the knowledge that area A is woodland and area B is woodland, and no area A is woodland and area B is woodland, and no information on any other area, the system might information on any other area, the system might infer that all areas are woodlandinfer that all areas are woodland

23/4/2023/4/20 1616

6. ISSUE

knowledge based systems have been only knowledge based systems have been only moderately successful in areas where problems moderately successful in areas where problems are relatively straightforward, e.g. medical are relatively straightforward, e.g. medical diagnosis diagnosis

some of the most successful applications have some of the most successful applications have been for instruction been for instruction e.g. use of medical expert system to develop diagnostic e.g. use of medical expert system to develop diagnostic

skills - encouraging students to structure knowledge and skills - encouraging students to structure knowledge and process it systematically in response to a problem process it systematically in response to a problem

as precise, analytical models of knowledge and the as precise, analytical models of knowledge and the ways in which it is used, expert systems can enhance ways in which it is used, expert systems can enhance our understanding of human decision-making processes our understanding of human decision-making processes - e.g. how does a cartographer position labels on a map? - e.g. how does a cartographer position labels on a map?

23/4/2023/4/20 1717

6. ISSUE

several factors may impede greater use: several factors may impede greater use: high cost of developing system - building the knowledge high cost of developing system - building the knowledge

base base uniqueness of every application uniqueness of every application dynamic nature of knowledge - knowledge base is not static dynamic nature of knowledge - knowledge base is not static inadequacy of alternatives for knowledge representation - inadequacy of alternatives for knowledge representation -

few examples fit precisely within any one form, e.g. few examples fit precisely within any one form, e.g. production rules production rules

unwillingness to trust the decisions of a machine (no unwillingness to trust the decisions of a machine (no "bedside manner") "bedside manner")

response time deteriorates rapidly as knowledge base grows response time deteriorates rapidly as knowledge base grows most knowledge is "fuzzy" or uncertain - system must return most knowledge is "fuzzy" or uncertain - system must return

many possible answers to a problem - few problems have a many possible answers to a problem - few problems have a precise, single answer - technical difficulties of representing precise, single answer - technical difficulties of representing and processing fuzzy knowledge and processing fuzzy knowledge

poor design of user interface - not "user friendly" poor design of user interface - not "user friendly" user often wants the reasoning behind a decision, not just user often wants the reasoning behind a decision, not just

the decision itself the decision itself