Abstract :
The volume of data being generated nowadays is increasing at phenomenal rate. Extracting useful knowledge from such data collections is an important and challenging issue. A promising technique is the rough set theory, a new mathematical approach to data analysis based on classification of objects of interest into similarity classes, which are indiscernible with respect to some features. Rough set theory provides a formal framework for data mining. Feature selection is a preprocessing step in data mining, and it is very effective in reducing dimensionality, reducing irrelevant data, increasing learning accuracy and improving comprehensibility. In this paper, Quickreduct and the proposed Accelerated Quickreduct algorithms are first presented, followed by the C4.5 approach for rule induction. A comparative study is also performed with the proposed and Quickreduct algorithms. The experiments are carried out on public domain datasets available in UCI machine learning repository and the Human Immuno deficiency Virus (HIV) data set to analyze the performance study. Keywords: Data mining, Rough set, Feature selection, Quickreduct, Accelerated Quickreduct, Knowledge discovery.
