My Dating Site Thinks I'm a Loser: effects of personal photos and presentation intervals on perceptions of recommender systems

  • Authors:

  • Shailendra Rao;Tom Hurlbutt;Clifford Nass;Nundu JanakiRam

  • Affiliations:

  • Stanford University, Stanford, CA, USA;Stanford University - Intuit, Stanford, CA, USA;Stanford University, Stanford, CA, USA;Stanford University - Google, Stanford, CA, USA

  • Venue:
  • Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
  • Year:
  • 2009

Quantified Score

Hi-index 0.01

Visualization

Abstract

Receiving poor results from a personalized recommendation system is frustrating. When users try to compensate by putting on a "different face" and game the system, the results can be even more frustrating. This paper investigates how to improve the user experience of such systems by: 1) adding personal photos to increase self-awareness, and 2) providing recommendations interspersed with personal questions. A 2x2 web experiment (N=56) within the context of an online dating match recommendation system was used to assess these two effects. Displaying a person's photo stabilized both response strategies and liking of a recommender's poor suggestions. Additionally, presenting all of the results together at the end was less frustrating than spreading them out. These results demonstrate that simple interface design decisions can have profound effects on user behaviors and attitudes with personalized recommendation systems.