Contemplation of Machine Learning Algorithm under Distinct Datasets

Abstract

The paper analyses machine learning and statistics classification under supervised learning approach. The grail of this study is collation of Machine Learning on variegated datasets that are contemplated and compared on the basis of two important parameters viz. time and accuracy. For the purpose of study, 6 different supervised Machine Learning algorithms are implemented on datasets in WEKA tools by percentage splitting method in which 66% of the total data have been used to train the model and 34% is used to test the efficiency of the model. As an out-turn, general comparison is produced, guiding researchers in their area of interest and to choose best available algorithm depending on the dataset classification. The main objective of this paper is to provide the general comparison between variegated machine learning algorithms and which is best suitable and efficient for a particular situation as every algorithm varies according to the area of application and single algorithm is not surpassing in every scenario.