Chapter 11 Expert system architecture, representation of knowledge, Knowledge Acquisition, and Reasoning
Introduction to Knowledge Management Knowledge is information that is contextual, relevant, and actionable.
Introduction to Knowledge Management
Knowledge Engineering Knowledge Engineering is a Process of acquiring knowledge from experts and building knowledge base There are two Knowledge Engineering perspectives Narrow perspective Knowledge acquisition, representation, validation, inference, maintenance Broad perspective Process of developing and maintaining intelligent system © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Knowledge Engineering Process The KE steps are: Acquisition of knowledge General knowledge or metaknowledge From experts, books, documents, sensors, files Knowledge representation Organized knowledge Knowledge validation and verification Inferences Software designed to pass statistical sample data to generalizations Explanation and justification capabilities © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Knowledge Engineering Process Inference © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Knowledge Sources Knowledge Sources are Documented Undocumented Written, viewed, sensory, behavior Undocumented Memory Acquired from Human senses Machines © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Knowledge levels Knowledge Levels are: Shallow Deep Surface level Input-output Deep Problem solving Difficult to collect, validate Interactions between system components © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Knowledge categories Knowledge Categories Declarative Procedural Descriptive representation Procedural How things work under different circumstances How to use declarative knowledge Problem solving Metaknowledge Knowledge about knowledge © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Knowledge Engineers Professionals who elicit knowledge from experts Empathetic, patient Broad range of understanding, capabilities Integrate knowledge from various sources Creates and edits code Operates tools Build knowledge base Validates information Trains users © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Knowledge Engineers © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Knowledge Management Activities Knowledge management initiatives and activities Most knowledge management initiatives have one of three aims: To make knowledge visible To develop a knowledge-intensive culture To build a knowledge infrastructure
Elicitation Methods Manual Semiautomatic Automatic Based on interview Track reasoning process Observation Semiautomatic Build base with minimal help from knowledge engineer Allows execution of routine tasks with minimal expert input Automatic Minimal input from both expert and knowledge engineer © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Manual Methods Interviews Structured Unstructured Semistructured Goal-oriented Unstructured Complex domains Data unrelated and difficult to integrate Semistructured © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Manual Methods Process tracking: is a set of techniques to track reasoning processes Protocol analysis Document expert’s decision-making Think aloud process Observation Motor movements Eye movements © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Manual Methods Case analysis Critical incident User discussions Expert commentary Graphs and conceptual models Brainstorming Prototyping Multidimensional scaling for distance matrix Clustering of elements Iterative performance review © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Semiautomatic Methods Repertory grid analysis Personal construct theory Organized, perceptual model of expert’s knowledge Expert identifies domain objects and their attributes Expert determines characteristics (traits) and opposites for each attribute Expert distinguishes between objects using numerica scale e.g. Create a grid Expert transfer system (ETS) Computer program that interviews experts to elicit information Rapid prototyping used to determine sufficiency of available knowledge © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Repertory grid analysis © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Semiautomatic Methods, continued Computer based tools features: Ability to add knowledge to base Ability to assess, refine knowledge Visual modeling for construction of domain Creation of decision trees and rules Ability to analyze information flows Integration tools © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Automatic Methods Knowledge discovery by computers Inductive learning from existing recognized cases Neural computing mimicking human brain Genetic algorithms using natural selection © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Multiple Experts Scenarios Approaches Experts contribute individually Primary expert’s information reviewed by secondary experts Small group decision Panels for verification and validation Approaches Consensus methods Analytic approaches Automation of process through software usage Decomposition © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Automated Knowledge Acquisition Rule Induction Training set with known outcomes Creates rules, for example: If income is $70,000 or more, approve the loan. Assesses new cases © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Automated Knowledge Acquisition Advantages Complex problem domain Builder can be expert Saves time, money
Automated Knowledge Acquisition Difficulties Rules may be difficult to understand Experts needed to select attributes Algorithms are based on search process that produces fewer questions Rule-based classification problems Allows few attributes Many examples needed Examples must be cleansed Limited to certainties Examples may be insufficient © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Automated Knowledge Acquisition Interactive induction Incrementally induced knowledge General models Object Network Based on interaction with expert interviews Computer supported Induction tables IF-THEN-ELSE rules © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Evaluation, Validation, Verification Dynamic activities Evaluation Assess system’s overall value Validation Compares system’s performance to expert’s Concordance and differences Verification Building and implementing system correctly Can be automated © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Representation of Knowledge Production Rules IF-THEN Independent part, combined with other pieces, to produce better result Model of human behavior Examples IF condition, THEN conclusion Conclusion, IF condition If condition, THEN conclusion1 (OR) ELSE conclusion2 © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Artificial Intelligence Rules Common types Knowledge rules Declares facts and relationships Stored in knowledge base Inference Given facts, advises how to proceed Part of inference engines Metarules © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Artificial Intelligence Rules Pros and cons Advantages Easy to understand, modify, maintain Explanations are easy to get. Rules are independent. Modification and maintenance are relatively easy. Uncertainty is easily combined with rules. Limitations Huge numbers may be required Designers may force knowledge into rule-based entities Systems may have search limitations; difficulties in evaluation © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Representation of Knowledge Semantic Networks Graphical depictions Nodes and links Hierarchical relationships between concepts Reflects inheritance e.g. does Sam need food? © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Representation of Knowledge Frames All knowledge about object Hierarchical structure allows for inheritance Allows for diagnosis of knowledge independence Object-oriented programming Knowledge organized by characteristics and attributes Slots Subslots/facets Parents are general attributes Instantiated to children Often combined with production rules © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Knowledge Relationship Representations Decision tables Spreadsheet format All possible attributes compared to conclusions Decision trees Nodes and links Knowledge diagramming Computational logic Propositional True/false statement Predicate logic Variable functions applied to components of statements © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Reasoning In Rule-Based Systems Inference Engine Algorithms Directs search of knowledge base Forward chaining Data driven Start with information, draw conclusions Backward chaining Goal driven Start with expectations, seek supporting evidence Inference/goal tree Schematic view of inference process AND/OR/NOT nodes Answers why and how Rule interpreter © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Explanation Facility Explanation facility is:The part of an ES that provides explanations has several purposes: Justifier Makes system more understandable Exposes shortcomings Explains situations that the user did not anticipate Satisfies user’s psychological and social needs Clarifies underlying assumptions Conducts sensitivity analysis Types Why How Journalism based Who, what, where, when, why, how Why not © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang
Generating Explanations Static explanation Preinsertion of text Dynamic explanation Reconstruction by rule evaluation Tracing records or line of reasoning Justification based on empirical associations Strategic use of metaknowledge © 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang