Credit apportionment scheme for rule based systems

Abstract EN

Credit apportionment scheme is the backbone of the performance of adaptive rule based systems.

The more cases the credit apportionment scheme can consider, the better is the overall system performance.

Currently rule based systems are used in various areas such as expert systems and machine learning which means that new rules to be generated and others to be eliminated.

Several credit apportionment schemes have been proposed and some of them are even used but still most of these schemes suffer from disability of distinguishing between good rules and bad rules.

Correct rules might be weakened because they are involved in an incorrect inference path and incorrect rules might be strengthen because they are involved in a correct conclusion.

We propose a credit apportionment scheme that is able to distinguish between good rules and bad rules based on their inference involvement.

The proposed algorithm is based on the overall system performance, rule weight, ratio of rule firing to the number of correct conclusions obtained using that rule, and ratio of rule firing to the number of incorrect conclusions obtained using that rule.