Building Expert Systems in PROLOG

ISBN-13: 9781461389132 / Angielski / Miękka / 2011 / 358 str.

Dennis Merritt

Opis

When I compare the books on expert systems in my library with the production expert systems I know of, I note that there are few good books on building expert systems in Prolog. Of course, the set of actual production systems is a little small for a valid statistical sample, at least at the time and place of this writing - here in Gennany, and in the first days of 1989. But there are at least some systems I have seen running in real life commercial and industrial environments, and not only at trade shows. I can observe the most impressive one in my immediate neighborhood. It is installed in the Telephone Shop of the Gennan Federal PTT near the Munich National Theater, and helps configure telephone systems and small PBXs for mostly private customers. It has a neat, graphical interface, and constructs and prices an individual telephone installation interactively before the very eyes of the customer. The hidden features of the system are even more impressive. It is part of an expert system network with a distributed knowledge base that will grow to about 150 installations in every Telephone Shop throughout Gennany. Each of them can be updated individually overnight via Teletex to present special offers or to adapt the selection process to the hardware supplies currently available at the local ware houses."

Spis treści

1 Introduction.- 1.1 Expert Systems.- 1.2 Expert System Features.- Goal-Driven Reasoning.- Uncertainty.- Data Driven Reasoning.- Data Representation.- User Interface.- Explanations.- 1.3 Sample Applications.- 1.4 Prolog.- 1.5 Assumptions.- 2 Using Prolog’s Inference Engine.- 2.1 The Bird Identification System.- Rule Formats.- Rules About Birds.- Rules for Hierarchical Relationships.- Rules for Other Relationships.- 2.2 User Interface.- Attribute Value pairs.- Asking the User.- Remembering the Answer.- Multi-Valued Answers.- Menus for the User.- Other Enhancements.- 2.3 A Simple Shell.- Command Loop.- A Tool for Non-Programmers.- 2.4 Summary.- Exercises.- 3 Backward Chaining with Uncertainty.- 3.1 Certainty Factors.- An Example.- Rule Uncertainty.- User Uncertainty.- Combining Certainties.- Properties of Certainty Factors.- 3.2 MYCIN’S Certainty Factors.- Determining Premise CF.- Combining Premise CF and Conclusion CF.- Premise Threshold CF.- Combining CFs.- 3.3 Rule Format.- 3.4 The Inference Engine.- Working Storage.- Find a Value for an Attribute.- Attribute Value Already Known.- Ask User for Attribute Value.- Deduce Attribute Value from Rules.- Negation.- 3.5 Making the Shell.- Starting the Inference.- 3.6 English-like Rules.- Exercises.- 4 Explanation.- Value of Explanations to the User.- Value of Explanations to the Developer.- Types of Explanation.- 4.1 Explanation in Clam.- Tracing.- How Explanations.- Why Questions.- 4.2 Native Prolog Systems.- Exercises.- 5 Forward Chaining.- 5.1 Production Systems.- 5.2 Using Oops.- 5.3 Implementation.- 5.4 Explanations for Oops.- 5.5 Enhancements.- 5.6 Rule Selection.- Generating the Conflict Set.- Time stamps.- 5.7 LEX.- Changes in the Rules.- Implementing LEX.- 5.8 MEA.- Exercises.- 6 Frames.- 6.1 The Code.- 6.2 Data Structure.- 6.3 The Manipulation Predicates.- 6.4 Using Frames.- 6.5 Summary.- Exercises.- 7 Integration.- 7.1 Foops (Frames and Oops).- Instances.- Rules for Frinsts.- Adding Prolog to Foops.- 7.2 Room Configuration.- Furniture Frames.- Frame Demons.- Initial Data.- Input Data.- The Rules.- Output Data.- 7.3 A Sample Run.- 7.4 Summary.- Exercises.- 8 Performance.- 8.1 Backward Chaining Indexes.- 8.2 Rete Match Algorithm.- Network Nodes.- Network Propagation.- Example of Network Propagation.- Performance Improvements.- 8.3 The Rete Graph Data Structures.- 8.4 Propagating Tokens.- 8.5 The Rule Compiler.- 8.6 Integration with Foops.- 8.7 Design Tradeoffs.- Exercises.- 9 User Interface.- 9.1 Object Oriented Window Interface.- 9.2 Developer’s Interface to Windows.- 9.3 High-Level Window Implementation.- Message Passing.- Inheritance.- 9.4 Low-Level Window Implementation.- Exercises.- 10 Two Hybrids.- 10.1 CVGEN.- 10.2 The Knowledge Base.- Rule for Parameters.- Rules for Derived Information.- Questions for the User.- Default Rules.- Rules for Edits.- Static Information.- 10.3 Inference Engine.- 10.4 Explanations.- 10.5 Environment.- 10.6 AIJMP.- 10.7 Summary.- Exercises.- 11 Prototyping.- 11.1 The Problem.- 11.2 The Sales Advisor Knowledge Base.- Qualifying.- Objectives — Benefits — Features.- Situation Analysis.- Competitive Analysis.- Miscellaneous Advice.- User Queries.- 11.3 The Inference Engine.- 11.4 User Interface.- 11.5 Summary.- Exercises.- 12 Rubik’s Cube.- 12.1 The Problem.- 12.2 The Cube.- 12.3 Rotation.- 12.4 High Level Rules.- 12.5 Improving the State.- 12.6 The Search.- 12.7 More Heuristics.- 12.8 User Interface.- 12.9 On the Limits of Machines.- Exercises.- Appendix A Native.- Sample Dialog.- Birds Knowledge Base.- Native Shell.- Appendix B Clam.- Sample Dialog.- Car Knowledge Base.- Clam.- Ldruls.- Appendix C Oops.- Sample Dialog.- Room Knowledge Base.- Oops.- Appendix D Foops.- Sample Dialog.- Room Knowledge Base (Foops).- Foops.- Appendix E Rete-Foops.- Rete Compiler and Runtime.- Appendix F Windows.- Window Demonstration.- Windows (abbreviated).- Appendix G Rubik.- Rubik.- Rubdata.- References.- Predicate Index.