Automatically building research reading lists

Michael D. Ekstrand, Praveen Kannan, James A Stemper, John T. Butler, Joseph A. Konstan, John T. Riedl

Research output: Chapter in Book/Report/Conference proceedingConference contribution

71 Scopus citations


All new researchers face the daunting task of familiarizing themselves with the existing body of research literature in their respective fields. Recommender algorithms could aid in preparing these lists, but most current algorithms do not understand how to rate the importance of a paper within the literature, which might limit their effectiveness in this domain. We explore several methods for augmenting exist-ing collaborative and content-based filtering algorithms with measures of the inuence of a paper within the web of cita-tions. We measure inuence using well-known algorithms, such as HITS and PageRank, for measuring a node's im-portance in a graph. Among these augmentation methods is a novel method for using importance scores to inuence collaborative filtering. We present a task-centered evalua-tion, including both an ofiine analysis and a user study, of the performance of the algorithms. Results from these stud-ies indicate that collaborative filtering outperforms content-based approaches for generating introductory reading lists.

Original languageEnglish (US)
Title of host publicationRecSys'10 - Proceedings of the 4th ACM Conference on Recommender Systems
Number of pages8
StatePublished - 2010
Event4th ACM Recommender Systems Conference, RecSys 2010 - Barcelona, Spain
Duration: Sep 26 2010Sep 30 2010

Publication series

NameRecSys'10 - Proceedings of the 4th ACM Conference on Recommender Systems


Other4th ACM Recommender Systems Conference, RecSys 2010


  • Citation web
  • Collaborative filtering
  • Digital libraries
  • Recom-mender systems
  • User study


Dive into the research topics of 'Automatically building research reading lists'. Together they form a unique fingerprint.

Cite this