Content Based Call for Papers Recommendation to Researchers

Abstract

Call for papers is an invitation to researchers for paper publication. Finding relevant conference for paper publication is important as it has a direct impact on researchers' profile, papers' acceptance and future citation of research work. However finding relevant conference is a time consuming job for researchers due to increasing number of conferences on daily basis. To address these issues, various Recommender Systems (RS) have been developed. Most of them are based on collaborative approaches which exploit users' preferences for recommendations, however such RS suffer from cold start problem. On the other hand, systems based on content based approaches uses items' features for recommendations, however these RS face various problems including limited content analysis and irrelevant recommendations. We addressed these issues by developing a content based CFP (Call for Papers) recommender system using selected features that can reflect researchers' preferences. These features include title, abstract, keywords, cited papers' titles and cited events. Experimental results show that the proposed system solve problems that traditional CFP recommender systems face and produce quality recommendation results.