Enhancement of web proxy caching using discriminative multinomial Na

Abstract

The discriminative multinomial Naive Bayes classifier has been used in the field of web data classification. In this study, we attempt to improve the performance of the web cache replacement policies such as SIZE and GDSF by applying the machine learning technique for enhancing the performance of the web proxy server. In the first part of this paper supervised learning method discriminative multinomial Naive Bayes (DMNB) classifier is used to train and classify the web log data and forecast the classes of web objects to be revisited or not. In the second part, a discriminative multinomial Naive Bayes (DMNB) classifier is incorporated with traditional web proxy caching policies to form novel caching approaches known as DMNB-SIZE and DMNB-GDSF. This proposed method significantly improves the performances of SIZE and GDSF in terms of cache hit and byte hit ratio respectively.