Comparison of Data Mining Classification Techniques Naïve Bayes, J48, IBK And SMO Using Commercial Dataset

Authors

Yasser Mahmood Hussain
General directorate of education in Al-Anbar- Iraq.
Dr. Murtada Khalafallah Elbashir
Jouf University, Sakaka, KSA, Saudi Arabia.
Dr. Mohammed Babiker Ali Mohammed
Dean of faculty of computer science and information technology – University of Al Butana, Butana.

Abstract

Commercial and mall datasets need to using the most known data mining techniques to classify goods and to predict the user requirements , behavior and company strategies. Classification applied to the financial dataset used which is called Wholesale customers dataset which can be collected by any other company or mall be filling the attributes with the information required which is contains 440 instances of 8 attributes, it refers to clients of a wholesale distributor. It includes the annual spending in monetary units (m.u.) on diverse product categories. In this paper many classification algorithm is been tested with the dataset used and compared by using many characteristics to explain the best algorithm used for such data classification. In order to classify or predict commercial data many algorithms used by companies and malls, these algorithms helps the owners and management of these companies for digging the data and analyze it and then extract meaningful data which can be used to take decisions related to it. Four data mining classification algorithms is been used Naïve Bayes, J48, IBK and SMO and results is discussed in details.