文章基本信息

标题：Aggregating Mechanism for Decisive the Ultimate Classification from the Ensemble to Boost Accuracy Rates
本地全文：下载
作者：A.Rama ; Gayathiri
期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2015
卷号：3
期号：3
出版社：S&S Publications
摘要：Classifier ensembles are used with success to boost accuracy rates of the underlying classificationmechanisms. Through the utilization of collective classifications, it becomes doable to attain lower error rates inclassification than by employing a single classifier instance. Ensembles area unit most frequently used withcollections of call trees or neural networks because of their higher rates of error once used severally. during thispaper, we are going to contemplate a novel implementation of a classifier ensemble that utilizes kNN classifiers.every categoryifier is ready-made to police investigation membership in a very specific class employing a best setchoice method for variables. This provides the range required to with success implement associate ensemble.associate aggregating mechanism for decisive the ultimate classification from the ensemble is conferred and testedagainst many documented datasets.
关键词：k Nearest Neighbor; Classifier Ensembles; Forward set choice