AN EMPIRICAL-STUDY OF DOMAIN KNOWLEDGE AND ITS BENEFITS TO SUBSTRUCTURE DISCOVERY

Discovering repetitive, interesting, and functional substructures in a structural database improves the ability to interpret and compress th e data. However, scientists working with a database in their area of e xpertise often search for predetermined types of structures or for str uctures exhibiting characteristics specific to the domain. This paper presents a method for guiding the discovery process with domain-specif ic knowledge. In this paper, the SUBDUE discovery system is used to ev aluate the benefits of using domain knowledge to guide the discovery p rocess. Domain knowledge is incorporated into SUBDUE following a singl e general methodology to guide the discovery process. Results show tha t domain-specific knowledge improves the search for substructures that are useful to the domain and leads to greater compression of the data . To illustrate these benefits, examples and experiments from the comp uter programming, computer-aided design circuit, and artificially gene rated domains are presented.