Intelligent Web caching using Adaptive Regression Trees, Splines, Random Forests and Tree Net

Abstract

Web caching is a technology for improving network traffic on the internet. It is a temporary storage of Web objects (such as HTML documents) for later retrieval. There are three significant advantages to Web caching; reduced bandwidth consumption, reduced server load, and reduced latency. These rewards have made the Web less expensive with better performance. The aim of this research is to introduce advanced machine learning approaches for Web caching to decide either to cache or not to the cache server, which could be modelled as a classification problem. The challenges include identifying attributes ranking and significant improvements in the classification accuracy. Four methods are employed in this research; Classification and Regression Trees (CART), Multivariate Adaptive Regression Splines (MARS), Random Forest (RF) and TreeNet (TN) are used for classification on Web caching. The experimental results reveal that CART performed extremely well in classifying Web objects from the existing log data and an excellent attribute to consider for an accomplishment of Web cache performance enhancement.