Using SQL in CBR for Similarity Retrieval: The Case of the TQF Advisory System

Abstract

Case-Base Reasoning (CBR) is a methodology that stands out as one the most useful artificial intelligence techniques. The essential idea of CBR is to answer user�s queries by comparing them with problems in the case base that have been solved and determine the most similar one. Case retrieval is a procedure that a retrieval algorithm recovers the most similar cases to the present problem. Essential data for corporate exercises is stored in big databases. While conventional database management systems offer restricted query flexibility, systems that can create similar based queries, for example, those found is case-based reasoning research, would improve the utility of data resources. This paper explains a strategy for building case-based systems utilizing a conventional relational database (RDB). The core of the algorithm is a new way to deal with similarity computing in which database queries form similarities. The implementation utilizes Structured Query Language (SQL) to accomplish the matching case report, and permits similarity base database retrieval.



Author Information
Putsadee Pornohol, Phuket Rajabhat University, Thailand
Suphamit Chittayasothorn, King Mongkut's Institute of Technology Ladkrabang, Thailand

Paper Information
Conference: ACSET2016
Stream: Education, Technology and Society: Technologies, Knowledge Creation and Access

This paper is part of the ACSET2016 Conference Proceedings (View)
Full Paper
View / Download the full paper in a new tab/window

Posted by amp21