Utilizing grammatical relations to improve recall and precision in a textual database

This paper explores the possibility of applying a model of natural language communication to textual databases and the WWW. The model, called Database Semantics (DBS), is designed as an artificial cognitive agent with a hearer mode, a think mode, and a speaker mode. For the application at hand, the hearer mode is used for (i) parsing language data into sets of proplets, defined as non-recursive feature structures, which are stored in a content-addressable memory called Word Bank, and (ii) for parsing the user query into a DBS schema employed for retrieval. The think mode is used to expand the primary data activated by the query to a wider range of relevant secondary and tertiary data. The speaker mode is used to realize the data retrieved in the natural language of the query. It is argued that DBS schemata based on the grammatical relations of functor-argument and coordination structure help to improve recall and precision.