Paper Title
RT And J48 Classification Models for Predicting Abnormal Web Multimedia Data

Now a days, the ‘Data Science Engineering’ becoming emerging trend to discover knowledge from web multimedia such as- YouTube multimedia, Yahoo Screen, Face Book multimedia etc. Petabytes of web multimedia are being shared on social websites and are being used by the trillions of users all over the world. Recently, discovering outliers among large scale web multimedia have attracted attention of many web multimedia mining researchers. There are plenty of outliers/abnormal multimedia exists in different category of web multimedia. The task of classifying and prediction of web multimedia as- normal and abnormal have gained vital research aspect in the area of Web Mining Research. Hence, we propose novel techniques to predict outliers from the web multimedia dataset based on their metadata objects using data mining Decision Tree algorithms such as Random Tree (RT) and J48 Tree algorithms. The results of Random Tree and J48 Tree classification models are analyzed and compared as a strategy in the process of knowledge discovery from web multimedia. Keywords - Outliers, Decision Tree, J48 Tree, Web Multimedia Outliers, Prediction, Knowledge Discovery.