Citation Network and Author based Search Pattern for Article Recommendation

Abstract

With the rapid growth in the information technology and the growing amount of the scholarly data, the recommendation techniques are becoming more popular. The main aim of the recommender system is to automatically suggest the users with the items of the potential interest. In the recent years they are being employed in the digital libraries for the researcher to get the required article. But this search tool gives the list of the relevant articles based in the keyword based queries. If any researchers have the same set of keywords then the obtained result would also be same. It is not feasible to describe the searchers need depending on the several limited keywords. In the proposed system, a novel recommendation method is being introduced in which it incorporates common author relations between the articles to help generate better recommendations for relevant target researcher. In this system the citation network is being constructed in order to give the relevant output to the researcher. Also the hybrid framework is being considered for the implementation. In order to determine the relevant researchers with author-based search patterns, the ratio of pairwise articles with common author relations and the ratio of the most frequently appeared author.