Title :
A graphical expert system for microfossil identification for use in the petroleum industry
Author_Institution :
BP Res. Int., Sunbury-on-Thames, UK
Abstract :
Describes the design and development of a graphical expert system for the identification of microfossils. The main components of the system are a graphical expert system shell and a set of knowledge bases. The process of knowledge elicitation is highly structured, and partly automated for maximum efficiency. The elicited knowledge for microfossils is stored in specially designed metafiles which are easily modified using the automated elicitation process. A knowledge base building tool generates knowledge bases automatically from the metafiles which makes the expert system easy to construct and extend. The graphical expert system shell displays generic pictures of microfossils and their attributes. The user compares these generic pictures to the actual observation and selects the most appropriate values in each case. On the basis of this information the expert system suggests possible fossils. Combining extensive domain knowledge with a graphical interface makes this system powerful, flexible and effective
Keywords :
computer graphics; expert systems; petroleum industry; user interfaces; attributes; automated elicitation; domain knowledge; generic pictures; graphical expert system; graphical interface; knowledge base building tool; knowledge bases; knowledge elicitation; metafiles; microfossil identification; petroleum industry;
Conference_Titel :
AI in the User Interface, IEE Colloquium on
Conference_Location :
London