FLY: Venue Recommendation using Limited Context

Recommendation of publication venues is very much the need of the hour. There is a sea of options for research conferences and journals. As a result, it is often found that researchers are not even aware of many publication venues in their research area. Existing work uses information, such as all the references in a given paper, past publication venues of the author(s), co-author relationships, and the paper content to do venue recommendation. However, in this paper, we propose a system that uses very limited context, namely the paper title, abstract, and a venue network, which is constructed using only a small subset of authors' publication history, to do venue recommendation. Our venue recommendation system, FLY, gives 30% higher accuracy compared to the current state-of-the-art limited context venue recommendation system.