ITS-An Intelligent Decision Making System


Question: Can we make a decision better than that made by human experts? Yes, we can. ITS is designed to do this job. ITS is an integration of a database mining system and an expert system. The former one mines databases for knowledge and creates the knowledge base (KB); the latter one reads and infers the KB directly and generates the decision.

An expert system makes decisions based on the knowledge stored in the KB. If the knowledge stored is better than human experts' knowledge, then the decisions made will be better than those made by human experts.

A database mining system can mine databases for knowledge that is unknown to human experts. Storing the discovered knowledge into a KB not only can save time and cost in the KB creation phase, but also can guarantee the better decision to be made in the decision making phase.

The table-formed output of the database mining system Dbm can be served as the input KB of the expert system Adm. Because Adm can read and infer the table-formed KB directly, no format transform is required. The connection between these two systems is smooth and seamless, and hence the process performs with high efficiency.

The figure shows a heart disease KB table generated by Dbm based on a heart disease database. The results are unknown before. Once a new patient's record is entered to a patient database, a non-query Adm inference process will start to infer the KB using the new patient's data, and an inference result will be generated at once (no query happens). For instance, the patient's data include that the chest pain type = 4, the resting EKG result = 2, the Thal. = Reverse defect, etc. The inference result, "the patient will suffer from heart attack, and the certainty factor is 97% (by rule 1)" will be generated at once.

The price of ITS for VisualAge in Windows is $995.00. For special requirements, we can customize the system for you.

MicroExpert
1928 Interface Lane, #104
Charlotte, NC, 28262
Phone: 1-704-597-5627
FAX: 1-704-597-5627
E-mail: hmchang@mindspring.com


[ Catalog Home ]