Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

CORE Recommender: a plug in suggesting open access content


Published on

Presented at Repositories Fringe August 2017, Edinburgh.

Published in: Education
  • Be the first to comment

  • Be the first to like this

CORE Recommender: a plug in suggesting open access content

  1. 1. CORE Recommender: a plug in suggesting open access content Nancy Pontika, Lucas Anastasiou, Aristotelis Charalampous, Matteo Cancellieri, Samuel Pearce and Petr Knoth CORE Knowledge Media institute, The Open University
  2. 2. Discoverability of content in repositories • Criticism of repositories (Salo, 2008; Konkiel, 2012; Acharya, 2015) “[The institutional repository] is like a roach motel. Data goes in, but it doesn’t come out.” (Salo, 2008)
  3. 3. By integrating recommender into repositories, we can increase the number of incoming links to relevant resources, thereby increasing their accessibility. Why is it needed?
  4. 4. Source: CORE Services
  5. 5. Recommendations as a service CORE provides a non-personalised recommendation system for articles from across the global network of repositories. • 8 million full texts • 77 million metadata records • 2683 repositories
  6. 6. What effect can it have? • Increase accessibility means more engagement, higher Click-Through Rate (CTR) • Twice as often people access resources on CORE via its recommender system than via search.
  7. 7. Recommendable items in the CORE recommender
  8. 8. Source: CORE Recommender installations (and more to come…)
  9. 9. How does the CORE recommender work? • Currently an article-article recommender system • Preprocessing prior to the recommendations: enrichment with e.g. identifiers. document type and citation data • Similarity measure/ranking function • Post-filtering using record quality • Feedback (crowdsourcing a black list)
  10. 10. What the CORE recommender is… • Use of full text rather than just abstracts • Ensure open access availability of recommended articles • Free service • Integration with repositories and availability over API Recommender paper:
  11. 11. What the CORE recommender is not… What the CORE recommender is not! • It is not always right! But which recommendation system is? What the CORE recommender does not do? • Compare the “quality” of the recommended articles with the “quality” of the initial paper