International Journal of Multidisciplinary Sciences and Engineering (IJMSE)
Social network sites make it possible to search large amounts of data for characteristic rules and patterns. If applied to monitoring data recorded on a real-time data or business in a network, they can be used to post the in network site database. In this paper, the authors present "Supervised learning" a method to cascade the decision tree learning methods to classifying into either family oriented, comedy, romantic and horror activities in a social network site. They can used to build any one of the decision tree such as (ID3, C4.5 and CART), here the decision tree on each dataset refines the decision boundaries by learning the subgroups within the database.