A distributed intelligent information system with natural language input for ad hoc knowledge discovery in databases
- Simon Fraser Univ., Burnaby, British Columbia (Canada)
A distributed information system is described which features a graphic user interface incorporating natural language input and which provides ad hoc knowledge discovery in relational databases. The system is comprised of multiple processes which communicate with each other over a network. The knowledge discovery process involves extracting generalizations from data using background knowledge in the form of concept hierarchies and a learning procedure based upon an attribute-oriented induction technique. The natural language understanding process is a parser based on Head-Driven Phrase Structure Grammar (HPSG), a modern lexicon-based grammar formalism better equipped than older rule-based approaches for handling the often idiosyncratic behavior of words. To generate semantic interpretations, the parser makes use of a process which orders logical access paths in unnormalized databases based on the strength of their dependency structures and on their efficiency of execution.
- OSTI ID:
- 382845
- Report Number(s):
- CONF-960154--; ISBN 0-9648731-8-4
- Country of Publication:
- United States
- Language:
- English
Similar Records
Interpretation of natural-language data base queries using optimization methods
Augmenting a database knowledge representation for natural language generation