Inductive Databases and Knowledge Scouts

George Mason University, Fairfax VA

Investigators

Abstract

The goal of this research project is to develop, implement, and test a methodology for building inductive databases (IDBs), which extend conventional databases by integrating in them inductive inference capabilities. Such capabilities allow a database to answer questions that require synthesizing plausible knowledge on the basis of the facts in the databases. The methodology integrates a database with a host of inductive inference operators that can be evoked automatically through scripts, called Knowledge Scouts, and expressed in a Knowledge Generation Language. Proposed ideas are tested experimentally on two datasets, one characterizing life changes in post-communist countries, and the second relating lifestyles with diseases. http://www.mli.gmu.edu