Towards a personalized, scalable, and exploratory academic recommendation service

  • Authors:
  • Onur Küçüktunç;Erik Saule;Kamer Kaya;Ümit V. Çatalyürek

  • Affiliations:
  • The Ohio State University;The Ohio State University;The Ohio State University;The Ohio State University

  • Venue:
  • Proceedings of the 2013 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Literature search is an integral part of the academic research. Academic recommendation services have been developed to help researchers with their literature search, many of which only provide a text-based search functionality. Such services are suitable for a first-level bibliographic search; however, they lack the benefits of today's recommendation engines. In this paper, we identify three important properties that an academic recommendation service could provide for better literature search: personalization, scalability, and exploratory search. With these objectives in mind, we present a web service called theadvisor which helps the users build a strong bibliography by extending the document set obtained after a first-level search. Along with an efficient and personalized recommendation algorithm, the service also features result diversification, relevance feedback, visualization for exploratory search. We explain the design criteria and rationale we employed to make the theadvisor a useful and scalable web service with a thorough evaluation.