SIRUP: Enhancing Serendipity In Recommendations via User Perceptions

Creating serendipity (i.e. “pleasant surprises for users”) is a primary goal of intelligent recommender systems. This project proposes an interdisciplinary approach to enhance the serendipity of TV recommendations that combines complementary knowledge from three disciplines – Computer Science, Language & Cognition and Communication Science. The project examines the “back-end” or algorithms behind serendipitous TV recommendations (Computer Science), the “front-end” or the actual display of these recommendations (Language & Cognition), and the “effect” on users’ perceptions and satisfaction (Communication Science). The project builds on an existing recommender system that combines (a) enriched content semantics with linked data and (b) broadcaster viewing data, developed by one of the applicants in the context of ViSTATV (http://vista-tv.eu). We seek to further enhance this system by adding insights about serendipity both from the perspective of Language & Cognition and Communication Science.

Scenario: Current TV recommender systems typically rely on genre or actors, e.g., recommendations referring to other “suspenseful genres” or “documentaries”. However, less obvious relationships between programs (e.g., relationships between actors and directors, or settings and locations) exist that can be found in linked data and exploited for more serendipitous recommendations. However, to date, little is known how these “linked data patterns” can be exploited to maximize the serendipity of TV recommendations. Accordingly, the present project seeks to fill this research gap. More specifically, we ask:

  • Back-end: What linked data patterns effectively maximize serendipity? And how should linked data patterns be ranked in order to maximize serendipity? (Computer Science)
  • Front-end: What discourse profile for recommendations enhances serendipity? What content and form of TV recommendations results in maximum serendipity? (Language & Cognition)
  • Effect: How do TV recommendations that differ in their underlying linked data patterns and portrayals affect user’s perceptions of serendipity and subsequent TV choice behavior? (Communication Science)

Check out the project’s website: http://sirup.wmprojects.nl/

Supervisors

  • Tilo Hartmann
    Department of Communication Science
    VU University Amsterdam
  • Allison Eden
    Department of Communication Science
    VU University Amsterdam
  • Gerard Steen
    Department of Language & Communication
    VU University Amsterdam
  • Lora Aroyo
    Department of Computer Science
    VU University Amsterdam
  • Paul Groth
    Department of Computer Science
    VU University Amsterdam

Students

  •  Britt Hoeksema
    MSc Communication Science
  • Esra Atescelik
    MSc Computer Science

vulogo_sm