Categorize the video server in P2P networks based on seasonal and normal popularity videos using machine learning approach

Abstract

There is a wide-ranging use of Peer-to-Peer (P2P) computing and applications in majority of the key areas of Engineering and Technology. Devoid of any centralized server, they can share their content since peers are linked with each other. This is the reason why P2P computing gives enhanced communication among peers. It is essential for the video server to maintain the data content link in cache memory so the cache memory sizes will be enlarged to a definite level and also the cache needs to be securely sustained by each and every peers. By utilizing the Machine Learning method, the proposed method centers its concentration on classifying the video server depending on seasonal and non seasonal popularity. Two supervised Machine Learning algorithms are utilized in this paper and are explained as follows. The Case-Based Reasoning algorithm is utilized in order to sort out well-liked videos and the Averaged One-Dependence Estimators (AODE) algorithm is utilized to sort out video server into seasonal and non-seasonal. The first algorithm is based on Retrieve, Reuse, Revise and Retain methods and the latter algorithm sorts out the video server into seasonal and non-seasonal based video servers. The work simulated by Java programming language.