Premium
An expert system for diagnosis of post‐harvest potato diseases 1
Author(s) -
ADAMS S. S.,
STEVENSON W. R.,
DELHOTAL P.,
FAYET J.
Publication year - 1990
Publication title -
eppo bulletin
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.327
H-Index - 36
eISSN - 1365-2338
pISSN - 0250-8052
DOI - 10.1111/j.1365-2338.1990.tb01217.x
Subject(s) - expert system , ibm , computer science , lisp , microcomputer , ibm pc compatible , medical diagnosis , knowledge base , preprocessor , artificial intelligence , software , programming language , medicine , pathology , telecommunications , chip , materials science , nanotechnology
The Institut National de la Recherche Agronomique (INRA, France) has developed a LISP‐based computer program for diagnosis of post‐harvest potato problems (diseases, insect/nematode damage, and environmental/physiological disorders) which has been modified for use in USA production areas. This expert system simulates the behavior of a diagnostic consultant in interaction with a user. The system is composed of a knowledge base (called TUBERS) and an expert system shell (called MIRA). TUBERS contains the collective diagnostic expertise (encoded in the form of if ‐ then rules) of interviewed specialists in plant pathology, entomology, and agronomy. MIRA performs the reasoning and problem‐solving tasks (including searching TUBERS for applicable rules) while directing questions and issuing diagnostic reports to the user. The user simply responds to system questions with ‘yes’/‘no’/‘I don't know’ or any number of multiple choice responses. The expert system currently considers 35 disorders and we have verified that it accurately diagnoses each of those disorders when supplied with an appropriate series of user responses to system questions about observable symptoms. The system operates under MS‐DOS on IBM PC/AT/XT compatible machines fitted with an add‐on CL 1000 memory board and under Xenix‐SCO on most 80286 and 80386‐based microcomputers.