INTERNATIONAL JOURNAL OF INNOVATIONS IN APPLIED SCIENCES & ENGINEERING

International Peer Reviewed (Refereed), Open Access Research Journal

E-ISSN:2454-9258 | P-ISSN:2454-809X | Estd Year: 2015

ABSTRACT


EMPLOYABILITY OF FEATURE SELECTION LINKED TO ENSEMBLE LEARNING IN THE ENHANCEMENT OF CLASSIFIER LEARNING

Jahnvi Gupta

Vol. 3, Jan-Dec 2017

Page Number: 122-126

Abstract:

One of the basic errands in information mining is grouping. It is a lot significant in characterization to accomplish the most remarkable accuracy. In the field of information mining, various classifiers are available for the characterization task. Every classification methods have their advantages and disadvantages. A few strategies function admirably with certain informational indexes, while different procedures function admirably with other informational indexes. There have been numerous strategies proposed for further developing arrangement accuracy. One such method is pre-processing, which helps in working on the nature of the information. Another technique is to consolidate the classifiers, which will, thus, further develop the classification accuracy. In this paper, an experimental examination has been done on different strategies for further improving system accuracy. One method includes determination, which will choose the best highlights from the accessible highlights in the informational collection. Another methodology is ensemble learning, which consolidates multiple classifiers to improve the accuracy of classification.

Back Download