A conceptual, case-relation representation of text for intelligent retrieval

This research demonstrates that intelligent, conceptual retrieval is possible. It is an attempt to move from contemporary IR towards intelligent retrieval by using natural language text analysis techniques. Intelligent retrieval systems should help the user find information by allowing him or her to develop a line of reasoning while searching. Search operations should take as little of the user's concentration as possible. Conceptual retrieval enables the user to find information about ideas that he or she cannot name but can describe. Such information can be found even when the stored text does not contain a nominalization. To achieve conceptual retrieval, a viable semantic representation had to be made of the text to be stored. The strength of our semantic representation results from the use of Harold Somers's grid of twenty-eight definitive deep cases. The grid is designed to answer the strongest criticisms of case by combining grammatical and semantic roles in each case. The cases have been developed beyond their original capacity, but the theoretical framework and the grid itself were kept intact. A knowledge base of contract law cases have been constructed. The principal argument of each case has been analyzed according to Stephen Toulmin's 'good reasons' argument model. John Sowa's conceptual graphs have been used as a near-FOL notation. In addition to the semantic representations of each argument, the knowledge base contains a lexicon of legal concepts and rules for semantic selection. The dissertation concludes with a retrieval demonstration using questions derived from cases following those represented in the knowledge base. LOG+, a frame matching algorithm by Mara Miezitis along with some proposed adaptations, is used. The demonstration focuses on pattern-matching among conceptual definitions using spreading activation. Semantic constraints facilitate inference within a type hierarchy. A case-law retrieval system would ideally provide the lawyer with conceptual access to cases and could free him or her to develop arguments. The use of deep cases for the representation of large texts makes conceptual retrieval possible. Employing inference to locate implicit information gives us desirable advantages over contemporary IR system designs.