What is it about?
We define a social network of researchers, where two researchers are connected if they co-authored a paper. Given an author and a set of keywords (which may define a research interest or future publication name), we explore the problem of giving a recommendation of potential co-authors for the author, for that specific keywords context. We examine our success by measuring how we were able to predict real world collaborations (tagged by their paper title), and show the methods proposed are effective.
Featured Image
Why is it important?
We define a problem variant, which to the best of our knowledge was not explored before (in the essence of "predicting" a tagged link). We examine and show what scores work well in reality, in a way which may help other potential explorations for recommending personas on a social graph within a textual context.
Perspectives
I hope researchers of social networks may find this interesting and eye-opening, as it includes real-world data and defines a novel approach to a newly defined problem variant.
Lior Ebel
Hebrew University of Jerusalem
Read the Original
This page is a summary of: Recommending collaborators using keywords, May 2013, ACM (Association for Computing Machinery),
DOI: 10.1145/2487788.2488091.
You can read the full text:
Contributors
The following have contributed to this page
