Towards predicting web searcher gaze position from mouse movements

  • Authors:
  • Qi Guo;Eugene Agichtein

  • Affiliations:
  • Emory University, Atlanta, GA, USA;Emory University, Atlanta, GA, USA

  • Venue:
  • CHI '10 Extended Abstracts on Human Factors in Computing Systems
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

A key problem in information retrieval is inferring the searcher's interest in the results, which can be used for implicit feedback, query suggestion, and result ranking and summarization. One important indicator of searcher interest is gaze position - that is, the results or the terms in a result listing where a searcher concentrates her attention. Capturing this information normally requires eye tracking equipment, which until now has limited the use of gaze-based feedback to the laboratory. While previous research has reported a correlation between mouse movement and gaze position, we are not aware of previous work on automatically inferring searcher's gaze position from mouse movement or similar interface interactions. In this paper, we report the first results on automatically inferring whether the searcher's gaze position is coordinated with the mouse position - a crucial step towards predicting the searcher gaze position from the computer mouse movements.