A paper recommendation system based on user's research interests

Abstract

Researchers and scientists read articles to improve their studies. Researchers spend too much time and struggle to find the suitable article they are looking for. The purpose of article recommendation system is to reduce the time they spend and present to them the related articles they are not aware of. Classic article recommendation systems do not consider the user's information, they show the same results in the same sort for each researcher. In this study, an article recommendation system that takes into consideration the researcher's work field and the publisher's previous articles is presented. One of the most important innovations of this work is the use of TF-IDF and Cosinus similarity to make article recommendations taking user's past articles into consideration. As a result of the work, the users have been recommended articles, and the method we present has proved more successful results compared to equivalent methods according to f-Measure criterion. © 2018 IEEE.