Acquiring Knowledge in Extended Hierarchical Censored Production Rules (EHCPRS) System

Abstract

EHCPRs system is a knowledge representation and reasoning system for representing common sense knowledge and reasoning with it. In such a system an EHCPR is used as a unit of knowledge for representing any universal concept. There are a number of EHCPRs at various levels of hierarchy of knowledge structure in the EHCPRs system, which results in a tree of EHCPRs. This EHCPRs tree has the capability of continuous growth through new added EHCPRs to it at proper place as well as to get refined continuously with time through improvement in the already acquired EHCPRs. The EHCPRs tree will become stronger in terms of strength of implication and richer in knowledge as time passes. This paper discusses different schemes for enhancing the intelligence, i.e., the knowledge base and the database in the EHCPRs system. By simple and general snippets of code, the EHCPRs system is able to acquire new pieces of knowledge and assimilate it properly in the already acquired knowledge base. The EHCPRs system dynamically restructures the EHCPRs tree in each learning phase by maintaining consistency and minimizing redundancy as well.