1. 1 Knowledge Acquisition, Representation and Organization Dr. Jeffrey M. Sta. Ines

2. 2 Knowledge Acquisition and Validation Knowledge Engineering

3. 3 n appropriately to construct and explain lines- of-reasoning n Art of building complex computer programs that represent and reason with knowledge of the world – (Feigenbaum and McCorduck [1983])

4. 4 n Narrow perspective: knowledge engineering deals with knowledge acquisition, representation, validation, inferencing, explanation and maintenance

5. 5 Knowledge Engineering Process Activities n Knowledge Acquisition n Knowledge Validation n Knowledge Representation n Inferencing n Explanation and Justification Decision Support Systems and Intelligent Systems, Efraim Turban and Jay E. Aronson 6th ed, Copyright 2001, Prentice Hall, Upper Saddle River, NJ

6. 6 Knowledge Engineering Process (Figure 11.1) Knowledge validation (test cases) Knowledge Representation Knowledge Acquisition Encoding Inferencing Sources of knowledge (experts, others) Explanation justification Knowledge base

7. 7 Scope of Knowledge n Knowledge acquisition is the extraction of knowledge from sources of expertise and its transfer to the knowledge base and sometimes to the inference engine n Knowledge is a collection of specialized facts, procedures and judgment rules

8. 8 Knowledge Sources n Documented (books, manuals, etc.) n Undocumented (in people's minds) – From people, from machines n Knowledge Acquisition from Databases n Knowledge Acquisition Via the Internet

9. 9 Knowledge Levels n Shallow knowledge (surface) n Deep knowledge n Can implement a computerized representation that is deeper than shallow knowledge n Special knowledge representation methods (semantic networks and frames) to allow the implementation of deeper-level reasoning (abstraction and analogy): important expert activity n Represent objects and processes of the domain of expertise at this level n Relationships among objects are important

10. 10 Major Categories of Knowledge n Declarative Knowledge n Procedural Knowledge n Metaknowledge

11. 11 Declarative Knowledge Descriptive Representation of Knowledge n Expressed in a factual statement n Shallow n Important in the initial stage of knowledge acquisition

12. 12 Procedural Knowledge n Considers the manner in which things work under different sets of circumstances – Includes step-by-step sequences and how-to types of instructions – May also include explanations – Involves automatic response to stimuli – May tell how to use declarative knowledge and how to make inferences

13. 13 n Descriptive knowledge relates to a specific object. Includes information about the meaning, roles, environment, resources, activities, associations and outcomes of the object n Procedural knowledge relates to the procedures employed in the problem-solving process

14. 14 Metaknowledge Knowledge about Knowledge Metaknowledge refers to knowledge about the operation of knowledge-based systems Its reasoning capabilities

15. 15 Knowledge Acquisition Difficulties Problems in Transferring Knowledge n Expressing Knowledge n Transfer to a Machine n Number of Participants n Structuring Knowledge

16. 16 n Experts may lack time or not cooperate n Testing and refining knowledge is complicated n Poorly defined methods for knowledge elicitation n System builders may collect knowledge from one source, but the relevant knowledge may be scattered across several sources n May collect documented knowledge rather than use experts n The knowledge collected may be incomplete n Difficult to recognize specific knowledge when mixed with irrelevant data n Experts may change their behavior when observed and/or interviewed n Problematic interpersonal communication between the knowledge engineer and the expert Other Reasons

17. 17 Overcoming the Difficulties n Knowledge acquisition tools with ways to decrease the representation mismatch between the human expert and the program (“learning by being told”) n Natural language processor to translate knowledge to a specific representation n Impacted by the role of the three major participants – Knowledge Engineer – Expert – End user

18. 18 n Critical – The ability and personality of the knowledge engineer – Must develop a positive relationship with the expert – The knowledge engineer must create the right impression n Computer-aided knowledge acquisition tools n Extensive integration of the acquisition efforts

19. 19 Knowledge Acquisition Methods: An Overview n Manual n Semiautomatic n Automatic (Computer Aided)

20. 20 Manual Methods - Structured Around Interviews n Process (Figure 11.4) n Interviewing n Tracking the Reasoning Process n Observing n Manual methods: slow, expensive and sometimes inaccurate

21. 21 Manual Methods of Knowledge Acquisition Elicitation Knowledge base Documented knowledge Experts Coding Knowledge engineer

22. 22 Semiautomatic Methods n Support Experts Directly (Figure 11.5) n Help Knowledge Engineers

23. 23 Expert-Driven Knowledge Acquisition Knowledge base Knowledge engineer Expert Coding Computer-aided (interactive) interviewing

24. 24 Automatic Methods n Expert’s and/or the knowledge engineer’s roles are minimized (or eliminated) n Induction Method (Figure 11.6)

25. 25 Induction-Driven Knowledge Acquisition Knowledge base Case histories and examples Induction system

26. 26 Knowledge Modeling n The knowledge model views knowledge acquisition as the construction of a model of problem- solving behavior-- a model in terms of knowledge instead of representations n Can reuse models across applications

27. 27 Interviews n Most Common Knowledge Acquisition: Face-to-face interviews n Interview Types – Unstructured (informal) – Semi-structured – Structured

28. 28 Unstructured Interviews n Most Common Variations – Talkthrough – Teachthrough – Readthrough

29. 29 n The knowledge engineer slowly learns about the problem n Then can build a representation of the knowledge n Knowledge acquisition involves – Uncovering important problem attributes – Making explicit the expert’s thought process

30. 30 Unstructured Interviews n Seldom provides complete or well-organized descriptions of cognitive processes because – The domains are generally complex – The experts usually find it very difficult to express some more important knowledge – Domain experts may interpret the lack of structure as requiring little preparation – Data acquired are often unrelated, exist at varying levels of complexity, and are difficult for the knowledge engineer to review, interpret and integrate – Few knowledge engineers can conduct an efficient unstructured interview

31. 31 Structured Interviews n Systematic goal-oriented process n Forces an organized communication between the knowledge engineer and the expert n Procedural Issues in Structuring an Interview n Interpersonal communication and analytical skills are important

32. 32 Interviews - Summary n Are important techniques n Must be planned carefully n Results must be verified and validated n Are sometimes replaced by tracking methods n Can supplement tracking or other knowledge acquisition methods

33. 33 Recommendation Before a knowledge engineer interviews the expert(s) 1. Interview a less knowledgeable (minor) expert – Helps the knowledge engineer Learn about the problem Learn its significance Learn about the expert(s) Learn who the users will be Understand the basic terminology Identify readable sources 2. Next read about the problem 3. Then, interview the expert(s) (much more effectively)

34. 34 Tracking Methods n Techniques that attempt to track the reasoning process of an expert n From cognitive psychology n Most common formal method: Protocol Analysis

35. 35 Protocol Analysis n Protocol : a record or documentation of the expert's step-by-step information processing and decision-making behavior n The expert performs a real task and verbalizes his/her thought process (think aloud)

36. 36 Observations and Other Manual Methods n Observations n Observe the Expert Work

37. 37 Other Manual Methods n Case analysis n Critical incident analysis n Discussions with the users n Commentaries n Conceptual graphs and models n Brainstorming n Prototyping n Multidimensional scaling n Johnson's hierarchical clustering n Performance review

38. 38 Expert-driven Methods n Knowledge Engineers Typically – Lack Knowledge About the Domain – Are Expensive – May Have Problems Communicating With Experts n Knowledge Acquisition May be Slow, Expensive and Unreliable n Can Experts Be Their Own Knowledge Engineers?

39. 39 Approaches to Expert-Driven Systems n Manual n Computer-Aided (Semiautomatic)

40. 40 Manual Method: Expert's Self-reports Problems with Experts’ Reports and Questionnaires 1. Requires the expert to act as knowledge engineer 2. Reports are biased 3. Experts often describe new and untested ideas and strategies 4. Experts lose interest rapidly 5. Experts must be proficient in flowcharting 6. Experts may forget certain knowledge 7. Experts are likely to be vague

41. 41 Benefits n May provide useful preliminary knowledge discovery and acquisition n Computer support can eliminate some limitations