Volume 12, Issue 4
|
|
Volume 12, Issue 4
|
|
|
|
| Jul/Aug 1998 | |
| Theme: Knowledge Based Systems |
|
To Volume 12, Issue 3
|
To Volume 12, Issue 5
|
||||||||
 |
 |
||||||||
|
Features
|
| Knowledge Management and Microsoft's Active Platform: Part 1 - Henry Seiler explains how to separate business logic from application code to evolve, extend, and integrate the client/server solution and Internet. |
| Modeling Methodology 5: Mathematical Programming Languages -- Will Dwinnell explores math-oriented fourth generation programming languages as modeling tools, focusing on four that are usually associated with other fields but are worth considering for AI-related applications. |
| Expert System vs. Procedural Language Development -- Kevin M. Kronfeld and Dr. Alan C. Tribble compare the performance of these tools, examining the relative merits of both development environments and the types of problems each is best suited to solve. |
| Dependency Tracking in Dynamic Knowledge Engineering - James Veitch explores how to move beyond the confines of expert system rule based reasoning to a "spreadsheet-like" model, with the use of numerical minimization techniques. |
| A Flex-based Expert System for Sewage Treatment Works Support - M.L.E. Dixon, P.W. Grant, L.G. Moseley, and Clive Spenser discuss the development of a rule-based solution and the value of knowledge in a large and complex company. |
| AI@Work - Attar Software, Decisioneering, and JB Systems share customer success stories using AI in applications from apple thinning in Tasmania to keeping Con Edison humming in New York City to VAR assessments in agriculture. |
| Regulars | ||
| Editorial | ||
| Secret Agent Man - Chatterbot Conspiracy | by Don Barker | |
| Intelligence Files - And the Beat Goes On | by David Blanchard | |
| AI and the Net - Harvesting Knowledge From the Internet | by Mary Kroening | |
| Product Updates ----------------------------> | 9 late breaking product announcements from around the world in the fields of: | |
| Case-Based Reasoning | Data Mining | |
| Expert Systems | Expert System Development Tools | |
| Genetic Algorithms | Internet | |
| Neural Networks | ||
| PC AI Buyer's Guide ----------------------> | Decision Support | Expert Systems |
| Expert System Development Tools | ||
| Product Service Guide - Provides access to information on an entire category of products | ||
| PC AI Blackboard - AI advertisers bulletin board | ||
|
Advertiser List for 12.4
|
||
| AAAI | Gordian Institute | Prolog Development Center |
| AI Developers Inc. | Harlequin | Quality Monitoring & Control |
| Acquired Intelligence | Intelligent Systems Report | Rule Machines Corporation |
| Amzi! Inc. | KDD '98 | Salford Systems |
| Applied Logic Systems | KnowledgeBroker Inc. | Soft Warehouse Inc. |
| ATTAR Software USA | Knowledge Tech. E-Commerce | Stellar Technology |
| BioComp Systems Inc. | Knowledge Tech Know. Mgmt | System Dynamics International Inc. |
| BotSpot | Laughing Bear Productions | The Haley Enterprise Inc. |
| California Scientific Software | Logic Programming Assoc Ltd | The MathWorks |
| Churchill Systems | Nestor Inc. | Ultrexx Corporation |
| Comdex Enterprise | NeuroDimension | Ward Systems Group Inc. |
| Decisioneering | Palisade Corporation | WizSoft Inc. |
| Franz Inc. . | PC AI | Xi Computer Corporation |
| Frontier GlobalCenter | Production Sys Tech Inc. |
| In the headiest days of AI, many researchers were so excited about their early successes that they made predictions that appear quite outrageous today. Their overly optimistic forecasts of HAL-like machines in the near-term were based on a flawed but intuitively apparent assumption. If they were making such startling progress in automating complex thinking tasks, like playing chess, then surely digitally replicating seemingly simpler tasks, like reading and understanding a newspaper, could be even more easily accomplished. | |
| However, it turned out that these fundamental tasks require a form of "common sense" reasoning that is much more difficult to emulate electronically thatn narrowly defined problems like playing chess. As a consequence, while little noticeable progress has been made at building machines capable of common sense reasoning, a number of signigficant advancements have been achieved in the area of expert systems, which deals with specifically bounded domains of knowledge. | |
| This issue features some of the most interesting and important developments in expert systems, with special attention to the tools and techniques that are rapidly moving the field into the next millennium. In "Knowledge Management and Microsoft's Active Platform: Part 1," Henry Seiler, a 15 year veteran of expert system design, explains how to separate busineess logic (rapidly changing rules) from application code to evolve, extend, and integrate the client/server solution and Internet. Learn about the next step in the evolution of object-orientation, the component-based model, and how it provides a robust environment to create reusable multi-language code. | |
| Kevin Kronfeld and Dr. Alan Tribble's "Expert System vs. Procedural Language Development" compares the performance of these tools on a real world application. The article points out the advantages and disadvantages of both development environments and the types of problems each is best suited to solve. "Dependency Tracking in Dynamic Knowledge Engineering," by James Veitch, explores how to move beyond the confines of expert system rule based reasoning to a "spreadsheet-like" model, with the use of numerical minimization techniques. This dependency tracking approach vastly expands the kind of knowledge "searches" that can be performed in a world of information overload. The Dixon, Grant, Moseley, and Spenser, article "A Flex-based Expert System for Sewage Treatment Works Support" illustrates how a rule-based solution can provide a highly useful "what -if" analysis of various sophisticated scenarios. | |
| Our resident columnists also comment on programs that have benefited greatly from the evolution of expert systems. In "Secret Agent Man," I discover a conspiracy to take over the world by a group of malcontent chatterbots, while Dennis Merritt, in "AI and the Net," reviews smart software for making Internet navigation much easier. David Blanchard wades through the recent corporate mergers to discuss changes in the AI industry landscape. | |
| Although expert systems remain narrowly confined and "brittle" outside their limited domains of knowledge, they represent one of AI's biggest hits and, who knows, some researcher may be closing in on the holy grail of common sense reasoning as you read this. | |
| Don Barker Senior Editor |
|
||||
|
|
|
|
|