A source independent framework for research paper recommendation

As the number of research papers available on the Web has increased enormously over the years, paper recommender systems have been proposed to help researchers on automatically finding works of interest. The main problem with the current approaches is that they assume that recommending algorithms are provided with a rich set of evidence (e.g., document collections, citations, profiles) which is normally not widely available. In this paper we propose a novel source independent framework for research paper recommendation. The framework requires as input only a single research paper and generates several potential queries by using terms in that paper, which are then submitted to existing Web information sources that hold research papers. Once a set of candidate papers for recommendation is generated, the framework applies content-based recommending algorithms to rank the candidates in order to recommend the ones most related to the input paper. This is done by using only publicly available metadata (i.e., title and abstract). We evaluate our proposed framework by performing an extensive experimentation in which we analyzed several strategies for query generation and several ranking strategies for paper recommendation. Our results show that good recommendations can be obtained with simple and low cost strategies.