Authors : Danasingh Asir Antony Gnana Singh, Subramanian Appavu Alias Balamurugan and Epiphany Jebamalar Leavline
Abstract: Decision making is practiced in every moment of individual life. The correct decision leads the humanity in right path towards obtaining prosperity and secures the life from the losses. The good decision making mainly relay accurate prediction. The prediction is practiced in all the emerging fields to make decision. In medical field, prediction supports to diagnose the disease in order to prescribe the correct medicine. In finance, prediction assists to predict the future demand and supply in order to satisfy the customer needs. In management, prediction helps to predict the profit and losses to lead the organization with maximum profitable. In engineering, the prediction supports to conduct the research and development activities. In management the prediction facilitates to predict the natural calamities to save and secure the life form the calamity. This prediction is carried out by the supervised learners known as supervised learners. The accuracy of these learners is determined by the significant variables presents in training dataset to train the learners. This study propose a novel algorithm namely Clustering with Variable Ranking and Selection algorithm (CVRS) to select most significant variable from the training dataset and remove the redundant and irrelevant variables form the training dataset. The performance of the proposed algorithm is compared with the six existing algorithms by four supervised learners. This proposed algorithm produces higher accuracy compared to other algorithms compared for the supervised learners.
Danasingh Asir Antony Gnana Singh, Subramanian Appavu Alias Balamurugan and Epiphany Jebamalar Leavline, 2014. Improving the Accuracy of the Supervised Learners using Unsupervised based Variable Selection. Asian Journal of Information Technology, 13: 530-537.