Development of a diagnostic expert system shell

• Main Authors: Nocon, Jessico E., Perez, Jonathan Ysidro A., Tierro, Frederick W., Zaide, Luis Enrique G.
• Format: text
• Language: English
• Published: Animo Repository 1989
• Subjects: Expert systems (Computer science); Electronic data processing; Computer Sciences
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/7339
• Institution: De La Salle University

The study is about the development of a diagnostic expert system shell (DEXSYS). DEXSYS allows users to create expert systems that are diagnostic in nature, such as medicine, automotive industry, business, science, etc. DEXSYS attempts to emulate through different applications the reasoning of human experts. DEXSYS was developed as an attempt to answer the inherent problems with regards to expert systems plus the limitations of current expert system building tool. In addition, like other A1 studies, DEXSYS is here not to replace human experts but rather to serve as an intelligent assistant. DEXSYS is not only a generalized inference engine, but it also covers the creation and modification of knowledge base in a much simpler manner. User interface by DEXSYS provides communication and explanation facility. The system made use of backward chaining for the inference mechanism, production rules for knowledge representation and Dempster Shafer Theory of Evidence for dealing with uncertainty. To be able to show the functions of DEXSYS a testbed application was presented in the field of medicine known as Gastroenterology.