Comparative Analysis of Selected Supervised Classification Algorithms.
No Thumbnail Available
Date
2015
Journal Title
Journal ISSN
Volume Title
Publisher
African Journal of Computing & ICT
Abstract
Information is not packaged in a standard easy-to-retrieve format. It is an underlying and usually subtle and misleading concept buried in
massive amounts of raw data. From the beginning of time it has been man’s common goal to make his life easier. The prevailing notion in
society is that wealth brings comfort and luxury, so it is not surprising that there has been so much work done on ways to sort large volume
of data. Over the year, there are various data mining techniques and used to sort large volume of data. This paper considers Classification
which is a supervised learning technique. Therefore the need to come up with the most efficient way to deal with voluminous data with
very little time frame has been one of the biggest challenges to the AI community. Hence, this paper presents a comparative analysis of
three classification algorithms namely; Decision Tree (J-48), Random Forest and Naïve Bayes. A 10-fold cross validation technique is used
for the performance evaluation of the classifiers on KDD’’99, VOTE and CREDIT datasets using WEKA (Waikato Environment for
Knowledge Analysis) tool. The experiment shows that the type of dataset determines which classifier is suitable.
Description
Keywords
Classification, Decision Tree (DT J-48), Random Forest (RF), Naïve Bayes (NB)