TV Program Popularity Prediction using Time Base

Abstract

The main goal is to the provide better recommendations system to TV viewers here the predictions are based on VOD streaming .and recommender systems have great achievements in the fields of VOD, (VIDEO ON DEMAND) .We have provide an effective recommended programs. Based on trending data .the rank wiil change based on time series.so that timely prediction of program popularity is of great value for content providers ,advertisers, and broadcast TV operators This information can be beneficial for operators in TV program purchasing decisions and can help advertisers formulate reasonable advertisement investment plans. Prediction models have been proposed based on video-on-demand (VOD) K spectral clustering and Kmedoids with Dynamic time warping are used for popularity prediction .and. file creation with multi grained is used for optimizing the streaming time. Cache replacement strategy is used to free up memory constraints. So it can speed up the streaming data.