On Ranking Discovered Rules of Data Mining by Data Envelopment Analysis: Some Models with Wider Applications

Abstract

The convergence of computing and communication has resulted in a society that feeds on information. There is exponentially increasing huge amount of information locked up in databases—information that is potentially important but has not yet been discovered or articulated (Whitten & Frank, 2005). Data mining, the extraction of implicit, previously unknown, and potentially useful information from data, can be viewed as a result of the natural evolution of Information Technology (IT). An evolutionary path has been passed in database field from data collection and database creation to data management, data analysis and understanding. According to Han & Camber (2001) the major reason that data mining has attracted a great deal of attention in information industry in recent years is due to the wide availability of huge amounts of data and the imminent need for turning such data into useful information and knowledge. The information and knowledge gained can be used for applications ranging from business management, production control, and market analysis, to engineering design and science exploration. In other words, in today’s business environment, it is essential to mine vast volumes of data for extracting patterns in order to support superior decision-making. Therefore, the importance of data mining is becoming increasingly obvious. Many data mining techniques have also been presented in various applications, such as association rule mining, sequential pattern mining, classification, clustering, and other statistical methods (Chen & Weng, 2008). Association rule mining is a widely recognized data mining method that determines consumer purchasing patterns in transaction databases. Many applications have used association rule mining techniques to discover useful information, including market basket analysis, product recommendation, web page pre-fetch, gene regulation pathways identification, medical record analysis, and so on (Chen & Weng, 2009). Extracting association rules has received considerable research attention and there are several efficient algorithms that cope with popular and computationally expensive task of association rule mining (Hipp et al., 2000). Using these algorithms, various rules may be obtained and only a small number of these rules may be selected for implementation due, at