A Specialized Evaluation and Comparison of Sample Data Mining Software

A Specialized Evaluation and Comparison of Sample Data Mining Software

John Wang (Montclair State University, USA), Xiaohua Hu (Drexel University, USA), Kimberly Hollister (Montclair State University, USA) and Dan Zhu (Iowa State University, USA)
DOI: 10.4018/978-1-60960-783-8.ch304

Chapter Preview

Top

Introduction

Based on the Knowledge Life Cycle model, four stages of knowledge creation, knowledge storage/retrieval, knowledge transfer, and knowledge application have been proposed by Alavi and Leidner (2001) and confirmed by Jennex (2006). “To be effective Knowledge Management Systems, KMS, must support the various knowledge management functions of knowledge capture, storage, search, retrieval, and use” (Jennex, 2006, p.3). Knowledge discovery is generally one of important stages or phases of KM. And while this incorporates identifying critical knowledge (this may also be what this stage is called), using data mining to aid in knowledge discovery is appropriate as being a useful KM tool.

Data mining is a promising tool that assists companies to uncover patterns hidden in their data. These patterns may be further used to forecast customer behavior, products and processes. It is important that managers who understand the business, the data, and the general nature of the analytical methods, are involved. Realistic expectation can yield rewarding results across a wide range of applications, from improving revenues to reducing costs (Davenport & Harris, 2007; Porter & Miller, 2001; Wang, 2008). It is crucial to properly collect and prepare the data, and to check the models against the real figures. The best model is often found after managers build models of several different types or by trying different technologies or algorithms. This alone demonstrates the active role managers play in the data mining or other knowledge management processes.

Selecting software is a practical yet very important problem for a company (James, Hakim, Chandras, King, & Variar, 2004). However, not enough attention is given to this critical task. Current literature is quite limited since selecting software is such a complex problem, due to many criteria and frequent technology changes (Elder IV & Abbott, 1998; Giraud-Carrier & Povel, 2003). Haughton, Deichmann, Eshghi, Sayek, Teebagy, & Topi (2003) generally reviewed several computer software packages for data mining including SPSS Clementine, XLMiner, Quadstone, GhostMiner, and SAS Enterprise Miner. Corral, Griffin, & Jennex (2005) examined the potential of knowledge management in data warehousing from an expert’s perspective. Jennex (2006) introduced technologies in support of knowledge management systems.

Firstly, this article will take a brief look at data mining today, through describing some of the opportunities, applications and available technologies. We will then discuss and analyze several of the most powerful data mining software tools available on the market today. Ultimately we will also attempt to provide an analytical analysis and comparison among the brands we have selected. Our selection is based, in part, on our own experience using data mining software as well as writing data mining code, SQL code and our work as relational database administrators. For our analytical comparison we will be using Expert Choice (Version 11) advanced decision support software (Wang, Hu, Hollister, & Zhu, 2008).

Complete Chapter List

Search this Book:
Reset