Scholarly publications have seen exponential growth in past decades, however, linear growth is estimated for the research topics. It suggests that numerous articles appear every day within a research field. Researchers invest their time and energy in keeping up-to-date with the current state-of-the-art. Research is a continuous process and it builds on the past work that has been done attributed through citations. Although, there are numerous reasons why a research article gets cited, as well as, its critics as to why citations should not be used to assess the value of current work. However, with the current information overload, it is not easy to keep abreast of all the published work. Researchers in 20th century would dig through all the available literature to find out the latest trends but the researcher of today has more stuff to read on a topic than their lifetime. They need access to current research as soon as it happens but the citation-count metrics, currently in practice, limit this approach. To use citation-based metrics, the articles must acquire a reasonable number of citations which can vary from field to field. Our main contribution is to use a heterogeneous network that includes the article, author and journal to recommend articles in a research field.
Import Ranking.py in your project
See Ranking.ipynb