knowledge representation. 2 knowledge representation and inference what is knowledge what is...

28
Knowledge Knowledge Representation Representation

Upload: allyssa-stoney

Post on 15-Dec-2015

240 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

Knowledge Knowledge RepresentationRepresentation

Page 2: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

2

KNOWLEDGE KNOWLEDGE REPRESENTATION AND REPRESENTATION AND

INFERENCEINFERENCE What is knowledgeWhat is knowledge What is knowledge representation What is knowledge representation

(KR)(KR) Knowledge representation languagesKnowledge representation languages Approaches to KRApproaches to KR

Semantic networksSemantic networks FramesFrames Predicate LogicPredicate Logic Production RulesProduction Rules

Page 3: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

3

Epistemology: Epistemology: theory of knowledgetheory of knowledge

What is knowledge? What is knowledge? Is genuine knowledge attainable at all? Is genuine knowledge attainable at all? What are the limits of knowledge? What are the limits of knowledge? From what faculties of the mind does From what faculties of the mind does

knowledge originate? knowledge originate? Which method should be used to Which method should be used to

obtain valid knowledge? obtain valid knowledge?

Page 4: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

4

The AI view of The AI view of knowledgeknowledge

Knowledge consists of models that attempt to represent the environment in such a way as to maximally simplify problem-solving.

It is assumed that no model can ever hope to capture all relevant information

Page 5: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

5

Basic assumption in Basic assumption in Artificial IntelligenceArtificial Intelligence

Intelligent behavior can be achieved through the manipulation of symbol structures (representing bits of knowledge).

This is based on the physical symbol system hypothesis, proposed by Newell and Simon 1976

Page 6: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

6

Main questionsMain questions

How we can represent knowledge as symbol structures ?

How we can use that knowledge to intelligently solve problems ?

Page 7: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

7

FeaturesFeatures

A knowledge representation (KR) : a substitute for the thing itself, used to enable an entity to determine consequences by thinking rather than acting

It is a set of ontological commitments, i.e. an answer to the question: In what terms should I think about the world?

Page 8: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

8

Knowledge Knowledge representation and other representation and other

fields of studyfields of study

Logic provides the formal structure and rules of inference

Ontology defines the kinds of things that exist in the application domain

Computation supports the applications that distinguish knowledge representation from pure philosophy

Page 9: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

9

Knowledge Knowledge Representation Representation

LanguagesLanguagesHigh level representation formalisms, that can in principle be implemented using programming languages.

should support inference.

inference - any way to get new expressions from old.

Page 10: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

10

RequirementsRequirements

Representational Adequacy:Representational Adequacy: should allow to represent the knowledge we need

Inferential Adequacy:Inferential Adequacy: ability to infer new knowledge from a basic set of facts

Inferential EfficiencyInferential Efficiency Clear Syntax and SemanticsClear Syntax and Semantics: : what the

allowable expressions are and what they mean

Naturalness:Naturalness: natural and easy to use

Page 11: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

11

Knowledge Representation and Knowledge Representation and Data structuresData structures

A KR is not a data A KR is not a data structurestructure

In KR: a correspondence between its constructs and things in the external world

KR - implemented by means of data structures

Page 12: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

12

Approaches to Knowledge Approaches to Knowledge Representation in AIRepresentation in AI

LogicLogic Semantic networks and FramesSemantic networks and Frames Production RulesProduction Rules

Declarative vs procedural knowledge

Declarative representations:

Page 13: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

13

First Order Predicate First Order Predicate LogicLogic

FOPL has a well defined syntax and semantics,

It is concerned with truth preserving inference.

Problems : time, beliefs and uncertainty are difficult to represent

Page 14: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

14

Semantic Nets and Semantic Nets and FramesFrames

• Represent factual knowledge about classes of objects and their properties• Not formal systems.• Basic inference mechanism: inheritance of properties

Problems: quantifiers, representing disjunction and negation

Page 15: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

15

Production rulesProduction rules Production systems: a set of if-then rules - typically state that if certain conditions hold, then some action should be taken.

If -then relation: If high_temperature then prescribe aspirin

Production systems use a working memory - represents the facts (as semantic nets or frames) that are currently believed to hold.

Page 16: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

16

Semantic NetsSemantic Nets

A semantic net is represented as a graph, where the nodes in the graph represent concepts, and the arcs represent binary relationships between concepts.

Nodes represent objects, attributes and values

Links represent attributes and relationships between nodes

Labels attached to links: the name of the corresponding attribute or relation

Page 17: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

17

animal

Is_a Is_a

reptile mammal head

Has part

Is_a

elephant

Clyde

Is_instance_of

Page 18: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

18

Types of RelationsTypes of Relations

depending on the application. (e.g. has_parts, likes, etc)

Important relations:

subclass / memberIs a / is instance of

Page 19: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

19

Inheritance of Inheritance of InformationInformation

More specific (sub)classes inherit (get) properties from more general (super)classes through

is_a / is_instance_of links

Example: Inferring facts not explicitly represented: Clyde has a head

Page 20: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

20

Representational and Representational and Inferential AdequacyInferential Adequacy

Problems with representing quantifiers, (such as ``every dog in town has bitten the constable'')

May have 2 arguments only

Cannot represent disjunction and negation

Page 21: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

21

ApplicationsApplications

ontologies, relational networks

Example:http://www.troubleshooters.com/tpromag/199907/_model.htm

Using Semantic Nets to Model Troubleshooting's Knowledge

Page 22: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

22

FramesFrames Proposed in 1968 by Marvin Minsky Proposed in 1968 by Marvin Minsky http://http://

web.media.mit.edu/~minskyweb.media.mit.edu/~minsky

All the information relevant to a particular All the information relevant to a particular concept is stored in a single complex concept is stored in a single complex entity, called a entity, called a frameframe..

Frames support Frames support inheritanceinheritance. . Frames can be viewed as a structural Frames can be viewed as a structural

representation of semantic nets.representation of semantic nets.

Page 23: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

23

ExampleExampleMammal subclass: Animal warm_blooded: yes

Elephant subclass: Mammal * colour: grey * size: large

Clyde instance: Elephant color: pink owner: Fred

Page 24: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

24

Components of a Frame Components of a Frame EntityEntity

Name - correspond to a node in a semantic net Attributes or slots filled with particular values

E.G. in the frame for Clyde, instance is the name of a slot, and elephant is the value of the slot.

• Names of slots correspond to the links in semantic nets

•Values of slots correspond to nodes.Hence each slot can be another

frame.

Page 25: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

25

ExampleExampleSize: instance: Slot single_valued: yes range: Size-set

Owner: instance: Slot single_valued: no range: Person

Fred: instance: Person occupation: Elephant-breeder

Page 26: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

26

Representational PowerRepresentational Power

Necessary attributes

Typical attributes (``*'' )

Type constraints and default values of slots,

Overriding values.

Slots and procedures: a slot may have a procedure to compute its value of the slot, if needed e.g. object area, given the size

Page 27: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

27

InheritanceInheritance

If a slot is not defined for a given frame, we look at the parent-class slot with the same name

Simple if single parent-class

several parent classes : multiple inheritance problem

(e.g., Clyde is both an elephant and a circus-animal)

Choose which parent to inherit from first.

Page 28: Knowledge Representation. 2 KNOWLEDGE REPRESENTATION AND INFERENCE What is knowledge What is knowledge What is knowledge representation (KR) What is knowledge

28

ApplicationsApplications

Classifying new instances of familiar Classifying new instances of familiar entities (objects/events/places/tasksentities (objects/events/places/tasks

Anticipating the attributes of such Anticipating the attributes of such instancesinstances

Inferring the presence and Inferring the presence and properties of their parts or properties of their parts or participantsparticipants