CPECSC 481 KnowledgeBased Systems Dr Franz J Kurfess
CPE/CSC 481: Knowledge-Based Systems Dr. Franz J. Kurfess Computer Science Department Cal Poly © 2005 Franz J. Kurfess Expert System Examples 1
Course Overview u Introduction u Knowledge u Semantic Nets, Frames, Logic u Reasoning u with Uncertainty Probability, Bayesian Decision Making u Expert u and Inference Predicate Logic, Inference Methods, Resolution u Reasoning u Representation System Design u CLIPS u Overview Concepts, Notation, Usage u Pattern u Matching Variables, Functions, Expressions, Constraints u Expert System Implementation u Salience, Rete Algorithm u Expert System Examples u Conclusions and Outlook ES Life Cycle © 2005 Franz J. Kurfess Expert System Examples 2
Outlook Knowledge-Based Systems u Motivation u Objectives u Intelligent u knowledge representation and reasoning for autonomous agents u Semantic u Agents u Knowledge u Management support for knowledge workers u Important Concepts and Terms u Chapter Summary Web reasoning with metadata and linked documents © 2005 Franz J. Kurfess Expert System Examples 3
Logistics u u Introductions Course Materials u u textbooks (see below) lecture notes u u u handouts Web page u u u Power. Point Slides will be available on my Web page http: //www. csc. calpoly. edu/~fkurfess Term Project Lab and Homework Assignments Exams Grading © 2005 Franz J. Kurfess Expert System Examples 4
Bridge-In © 2005 Franz J. Kurfess Expert System Examples 5
Pre-Test © 2005 Franz J. Kurfess Expert System Examples 6
Motivation u reasons to study the concepts and methods in the chapter u main advantages u potential benefits u understanding of the concepts and methods u relationships to other topics in the same or related courses © 2005 Franz J. Kurfess Expert System Examples 7
Objectives u regurgitate u basic facts and concepts u understand u u elementary methods more advanced methods scenarios and applications for those methods important characteristics v differences between methods, advantages, disadvantages, performance, typical scenarios u evaluate u application of methods to scenarios or tasks u apply u methods to simple problems © 2005 Franz J. Kurfess Expert System Examples 8
© 2005 Franz J. Kurfess Expert System Examples 10
Intelligent Agents u autonomous capabilities agents with knowledge-handling u knowledge representation and reasoning is often used for model building and decision making u exchange of knowledge among agents u relatively easy when agents use the same representation and reasoning method v still significant problems since their knowledge bases are not necessarily designed for exchange u use of specific knowledge exchange languages Knowledge Query and Manipulation Language (KQML) v ontology-based approaches (RDF, OWL, Semantic Web) v © 2005 Franz J. Kurfess Expert System Examples 11
Semantic Web u WWW u u u enhanced by meta-data and reasoning infrastructure XML as common base ontologies to define terms and relationships for models description logics as formal foundation Web services via e. g. Simple Object Access Protocol (SOAP) see the Scientific American article “The Semantic Web -- A new form of Web content that is meaningful to computers will unleash a revolution of new possibilities” by Tim Berners-Lee, James Hendler and Ora Lassila (May 2001), http: //www. sciam. com/print_version. cfm? article. ID=00048144 -10 D 21 C 70 -84 A 9809 EC 588 EF 21 © 2005 Franz J. Kurfess Expert System Examples 12
Semantic Web Examples u IRS Internet Reasoning Service ua Semantic Web services framework http: //kmi. open. ac. uk/projects/irs/ u Rule. ML u canonical Web language for rules using XML markup, formal semantics, and efficient implementations © 2005 Franz J. Kurfess Expert System Examples 13
IRS Internet Reasoning Service ua Semantic Web services framework available at http: //kmi. open. ac. uk/projects/irs/ © 2005 Franz J. Kurfess http: //kmi. open. ac. uk/projects/irs/ Expert System Examples 14
IRS Architecture ua server-client based approach IRS Server v IRS Publisher v IRS Client v © 2005 Franz J. Kurfess http: //kmi. open. ac. uk/projects/irs/ Expert System Examples 15
Rule. ML u covers u from the entire rule spectrum derivation rules to transformation rules to reaction rules u can specify u queries and inferences in Web ontologies u mappings between Web ontologies u dynamic Web behaviors of workflows, services, and agents u further information at the Rule Markup Initiative Web page http: //www. ruleml. org/ © 2005 Franz J. Kurfess Expert System Examples 16
Rule. ML Rules u rule u interoperation between industry standards v u such as JSR 94, SQL'99, OCL, BPMI, WSFL, XLang, XQuery, RQL, OWL, DAML-S, and ISO Prolog established systems v CLIPS, Jess, ILOG JRules, Blaze Advisor, Versata, MQWork. Flow, Biz. Talk, Savvion, etc. u modular u from and to other rule standards/systems u rules u u u Rule. ML specification and transformations can be stated in natural language in some formal notation in a combination of both © 2005 Franz J. Kurfess Expert System Examples 17
Rule. ML Example "The discount for a customer buying a product is 5. 0 percent if the customer is premium and the product is regular. " Note: This is one of several possible variations © 2005 Franz J. Kurfess <!-- Implication Rule 1 (permuted): Forward notation of _body and _head roles, similar to Production Systems (role permutation does not affect the semantics) --> <imp> <_body> <and> <atom> <_opr><rel>premium</rel></_opr> <var>customer</var> </atom> <_opr><rel>regular</rel></_opr> <var>product</var> </atom> </and> </_body> <_head> <atom> <_opr><rel>discount</rel></_opr> <var>customer</var> <var>product</var> <ind>5. 0 percent</ind> </atom> </_head> </imp> http: //www. ruleml. org/lib/discount-variations. ruleml Expert System Examples 18
Ontologies u definition u u u formal foundations, but still accessible for humans usually restricted to specific domains merge aspects of v v v u for u of terms and relationships dictionaries taxonomies and hierarchies semantic networks an introduction, see Ontology Development 101: A Guide to Creating Your First Ontology by Natalya F. Noy and Deborah L. Mc. Guinness, Stanford University, http: //www. ksl. stanford. edu/people/dlm/papers/ontology 101 -noymcguinness. html © 2005 Franz J. Kurfess Expert System Examples 19
Knowledge Management u support for knowledge workers u emphasis on knowledge representation and reasoning support for humans u knowledge © 2005 Franz J. Kurfess processing by computers is less important Expert System Examples 20
Chaotic vs. Systematic Knowledge Handling u chaotic u u u u heuristics unsound reasoning methods inconsistent knowledge jumping to conclusions ill-defined problems unclear boundaries of knowledge informal, continuous metareasoning © 2005 Franz J. Kurfess u systematic u u u u rules formal logic consistency proofs well-defined problems domain-specific knowledge expensive, distinct metareasoning Expert System Examples 21
Knowledge Fusion u integration of human-generated and machinegenerated knowledge u sometimes also used to indicate the integration of knowledge from different sources, or in different formats u can be both conceptually and technically very difficult u different “spirit” of the knowledge representation used u different terminology u different categorization criteria u different representation and processing mechanisms u ontologies attempt to build bridges u agreements © 2005 Franz J. Kurfess over basic terms, relationships Expert System Examples 22
© 2005 Franz J. Kurfess Expert System Examples 23
Questions © 2005 Franz J. Kurfess Expert System Examples 24
Figure Example © 2005 Franz J. Kurfess Expert System Examples 25
Post-Test © 2005 Franz J. Kurfess Expert System Examples 26
Important Concepts and Terms u u u u common-sense knowledge expert system (ES) expert system shell inference mechanism If-Then rules knowledge acquisition © 2005 Franz J. Kurfess u u u knowledge base knowledge-based system knowledge representation production rules reasoning rule Expert System Examples 28
Summary Outlook © 2005 Franz J. Kurfess Expert System Examples 29
© 2005 Franz J. Kurfess Expert System Examples 30
- Slides: 28