Study to Analyze the Performance of Supervised Machine Learning Classifiers
S. Magesh1, G. Chandar2
1S. Magesh, Professor, Department of Computer Science, Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences, Chennai (Tamil Nadu), India.
2G. Chandar, Research Scholar, Department of Computer Science, Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences, Chennai (Tamil Nadu), India.
Manuscript received on 26 April 2019 | Revised Manuscript received on 08 May 2019 | Manuscript Published on 17 May 2019 | PP: 439-442 | Volume-7 Issue-6S4 April 2019 | Retrieval Number: F10890476S419/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: On the rise exponentially, machine learning is adopted by various organizations across various domains to crunch petabytes of data to deliver more meaningful conclusions and predictions. With machine learning, it’s become quite possible to transform a wide variety of industries, in this complicated world. A category of machine learning algorithms called as classifiers, have found wide spread use in processing the large datasets efficiently. Classifiers are applied to big datasets through supervised learning methods. Though there exist huge number of classifier algorithms, it is important to apply some intelligence in careful selection of algorithms, to deliver excellent business intelligence. In order to select a more suitable classifier algorithm for a problem, it is important to analyze the performance of classifiers based on certain parameters and different datasets. In this paper, we analyzed variety of classifiers from the perspective of performance measures such as Precision, Recall, Mean Absolute Error, and Root Mean Square Error pertaining to two different datasets using WEKA. Also, in the conclusion we present the findings from analysis of performance of classifiers.
Keywords: Classifiers, Supervised, Machine Learning, Precision, Recall, WEKA.
Scope of the Article: Machine Learning