Rule Induction for Knowledge Acquisition in RKF

Introduction

The RKF Year 2 Challenge problem highlighted the difÞculty that Subject Matter Experts (SMEs) have in writing inference rules. During the evaluation period, the KRAKEN SMEs entered about Þve to seven rules each. One solution to this problem is to provide support for learning inference rules from examples rather than requiring domain experts to write the rules themselves.

We believe that this will provide another tool to accelerate the knowledge entry and knowledge base development. Learning from examples can help solve this particular prob- lem because it is easier to describe examples than to write rules. There may also be in- dependent sources of examples, perhaps in databases that donÕt support the expression of rules. Those examples could be imported and used as the input data to a rule induction algorithm.

This report describes the reÞnement and extension of a rule induction algorithm originally implemented in PowerLoom™ as part of the High Performance Knowledge Base (HPKB) project.

The full report is available in PDF. (25pp, 260kB)

Back to Paper List