Towards Integrating Hierarchical Censored Production Rule(HCPR) Based Systems and Neural Networks

Abstract

Over the past few years, researchers have successfully developed a number of systems that combine the strength of the symbolic and connectionist approaches to Artificial Intelligence. Most of the efforts have employed standard production rules, IF<condition> THEN <action> as underlying symbolic representation. In this paper we have suggested Variable Precision Neural Logic (VPNL) networks as an attempt towards integrating Hierarchical Censored Production Rule(HCPR) based system and neural networks. A HCPR has the form:

• Decision

  • (If precondition)

  • (Unless censor-conditions)

  • (Generality general-information)

  • (Specificity specific-information))

which can be made to exhibit variable precision in the reasoning such that both certainty of belief in a conclusion and its specificity may be controlled by the reasoning process. Also it is proposed how Dempster-Shafer uncertainty calculus can be incorporated into the inferencing process of VPNL networks. The proposed hybrid system would have numerous applications where decision must be taken in real time and with uncertain information – examples range from medical expert systems for operating rooms to domestic robots and with its capability of modeling wide range of human-decision making behavior, the system would have applications such as bond trading, currency option trading.