Enigma: A System That Learns Diagnostic Knowledge

This paper describes the results of an extensive experimentation aimed at assessing the concrete possibilities of automatically building a diagnostic expert system, to be used in-field in an industrial domain, by means of machine learning techniques. The system, ENIGMA, used in the experimentation, is an incremental version of the ML-SMART system, which acquires a network of first-order logic rules, starting from a set of classified examples and a domain theory. The application described in this paper has been selected, among several others, for its particular significance, both in terms of complexity of the solved problem and in terms of the obtained industrial benefits. The problem has been supplied by the ENICHEM Company and consists in discovering malfunctions in electromechanical apparata. ENIGMA's efficacy in acquiring sophisticated knowledge and handling complex structured examples is largely due to its underlying database management system, which supports the learning operators, defined at the abstract level, with a set of primitives, taken from the field of deductive databases. This database layer serves the additional goal of easing the interaction between the learning module and an information system, currently used at ENICHEM to directly stnn> the data measured in field. An expert system, MEPS, devoted to the same task, has also been manually developed. Then, a number of comparisons along different dimensions of the manual and automatic development process have been possible, allowing some practical indications to be suggested.