J. Martin and S. Oxman. Building Expert Systems: A Tutorial, Englewood Cliffs, N.J., Prentice-Hall, 1988.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....4. execute (file var) 5. exit (file var) 6. delete (file var) 7. modify owner (file vat ) 8. modify perm (file var) 9. hardlink (file var, file var) 10. renane (file var, file var) U. The Inference Engine USTAT s inference engine uses an event driven, forward chaining inference scheme [21]. The inference engine uses a structure called the inference engine table to keep track of all possible penetration scenarios. At any point in time, this table consists of snapshots of penetration scenario in stances (instantiations) that are not yet completed on the target system. Each row ....

J. Martin and S. Oxman. Building Expert Systems: A Tutorial, Englewood Cliffs, N.J., Prentice-Hall, 1988.

....which store the state assertions and the signature actions, respectively. The inference engine uses this information to match the actions of incoming USTAT audit records to the actions of state transition scenarios. USTAT s inference engine uses an event driven, forward chaining inference scheme [MO 88] The inference engine uses a structure called the inference engine table to keep track of all possible penetration scenarios. At any point in time, this table consists of snapshots of penetration scenario instances (instantiations) which are not yet completed on the target system. Each entry ....

J. Martin and S. Oxman, Building Expert Systems: A Tutorial, PrenticeHall, Englewood Cliffs, N.J., 1988.

.... The Fact base The Fact base Initializer The Fact base Updater The Rule base The State Description Table The Signature Action Table ffl The Inference Engine ffl The Decision Engine Except for the preprocessor, these components characterize a typical expert system. Martin and Oxman [Mart88] emphasize the modularity and independence of the expert system components in the following way. If an expert system component needs to be built or modified, it should not affect the other components. The major task of building an expert system will be the development of the knowledge base. STAT ....

....of an expert system in search of new information. The inference engine does not know what rules and facts should be or could be in the knowledge base. For any given inference step, the inference engine blindly uses all the relevant rules and facts that are available to it at that time. In [Mart88], Martin and Oxman list several problem solving activities that are supported by expert systems. STAT s activity is best described by monitoring, which is defined as follows. Monitoring involves observing and checking for some specific purpose. To monitor a system means to keep track of the ....

[Article contains additional citation context not shown here]

James Martin, Steven Oxman. Building Expert Systems: A Tutorial, Englewood Cliffs, N.J., Prentice-Hall, 1988.

....architectural distinctions) to rather abstract domain characteristics. For example, constraint propagation should be employed if subproblems interact, belief revision employed if search efficiency needs improving, etc. A similar picture is drawn in a recent 450 page tutorial on expert systems [19] Variations in system architecture are described in terms of the relative importance of the five main components of a system: inference engine, knowledge base, usersystem interface, explanation facility, and knowledge acquisition facility. Real time process control systems, for example, would ....

....a component list is much less problematic than an indirect specification, e.g. in terms of functional requirements or constraints. The latter would necessitate component selection as part of the configuration task. For DEC computers this is done by the separate expert system XSEL discussed in [19] which accepts customer wishes as input and delivers a component list as output. We now turn to problem solving knowledge as represented and used in XCON. It is explicitly expressed by rules and implicitly by the conflict resolution strategies of OPS5. The control thus achieved is remarkable. ....

[Article contains additional citation context not shown here]

Martin, J., Oxman, S.: Building Expert Systems - A Tutorial. Prentice Hall, Englewood Cliffs, New Jersey, 1988.

No context found.

James Martin and Steven Oxman. Building Expert Systems: A Tutorial. Prentice-Hall, 1988.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC