Recommendation system with collaborative social tagging exploration

Recommender system plays a significant role in reducing the information overload on the sites where users have searched and contented. Existing approaches which deals with such recommendation system apply collaborative filtering techniques to specify the most alike users whom they hope to make recommendations. Collaborative Filtering will significantly show better improvement with the enclosure of real data extraction from the suitable tagging system. In this paper, data from social tagging systems are extracted for every individual considering the correlations between users, items, and tag information. Tag information from users is the most decisive factor to predict the personalized suggestion for web users. Here, we rank the available content based tag information with the inclusion of temporal decay of users´ behavior over time and the centrality of every node in the network. Finally, we use the common preference metric for effective personalization. Results have been experimentally demonstrated with the empirical dataset MovieLens and provided the results as an alternative recommendation method with simplicity and efficiency.