WEB MULTIMEDIA CLASSIFICATION USING DT AND SVM MODELS

Abstract The ubiquity of the World Wide Web with media of all types such - ‘Entertainment’, ‘News and Politics’, ‘Sports’ etc, in recent years raised the attractiveness of web multimedia-video classification. It is usually desirable to combine evidence from different multimedia components such as audio, video, image and text etc. This work a novel approach to classify the multimedia contents based on internal metadata such asvideo bit rate kbps, maximum bit rate kbps, width pixels, height pixels etc. The proposed work adopts Decision Tree and Support Vector Machine (SVM) approach for classification process. The metadata of various components of web multimedia data are extracted and stored in a database for experimental purpose. The web multimedia data are labeled based on number of components present in the domain category of the multimedia are classified as KDD process.