The research focuses on improving current methods for learning to classify texts by incorporating knowledge from an expert domain model. The goal is automatically to classify the texts of legal opinions in terms of the factors that apply to the cases described. Factors -- stereotypical fact patterns tending to strengthen or weaken the underlying legal claims in a case, and their relations to legal issues -- are a kind of expert domain knowledge useful in legal argumentation. The program takes as inputs the raw texts of legal opinions and assigns as outputs the applicable factors. The program's training instances are drawn from a corpus of legal opinions whose textual descriptions of cases have been represented manually in terms of factors. The problem is hard because the language of the opinions is complex; the mere fact that an opinion discusses factors does not necessarily imply that those factors actually apply to the case. Starting with several existing inductive and statistical learning algorithms, this research assesses whether adding four different kinds of domain knowledge improves the algorithms' performance: (1) domain knowledge about factors and the legal issues to which they relate; (2) general information about the structure of legal opinions; (3) information about the statutes quoted in an opinion; (4) information about those cases cited in an opinion whose factors are known. The work also explores (a) how to combine inductive and analytical techniques to deal with small numbers of training instances and (b) how best to combine successful inductive, statistical, and knowledge-based methods. Using domain knowledge to guide automatic text classification integrates information retrieval, machine learning and AI knowledge-representation techniques, will help scale up case-based reasoning systems, and alleviate the problem of assessing the relevance of texts in increasingly large on-line databases.
