Abstract
This study aims to design a small, rule-based, expert system, which helps a user to estimate emplacement mechanisms of volcaniclastic rocks. To develop this system, an expert shell Intelligent Developer (Hyperpress Publishing Corp.) and a Macintosh computer (Apple Computer Inc.) were used and a system prototype Volcaniclast was made. Volcaniclast is composed of an inference engine, a rule base, a fact base, and a man-machine interface module. The rule base has 76 IF..THEN.. rules and the fact base has 11 facts. The interface module enables friendly, graphical, dialog-based environment for consultation.Although many unreal sub-answers are proposed, Volcaniclast gives a true answer with the highest certainty factor, when a typical set of field data are inputted. For this reason, Volcaniclast can at leastbe used for undergraduate students to learn fundamental knowledge in field volcanology.
