Proceedings of the 6th international conference on Intelligent user interfaces
A study on the effects of personalization and task information on implicit feedback performance
CIKM '06 Proceedings of the 15th ACM international conference on Information and knowledge management
Eye-mouse coordination patterns on web search results pages
CHI '08 Extended Abstracts on Human Factors in Computing Systems
Proceedings of the 18th international conference on World wide web
Towards predicting web searcher gaze position from mouse movements
CHI '10 Extended Abstracts on Human Factors in Computing Systems
No clicks, no problem: using cursor movements to understand and improve search
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
Detecting success in mobile search from interaction
Proceedings of the 34th international ACM SIGIR conference on Research and development in Information Retrieval
Reading and estimating gaze on smart phones
Proceedings of the Symposium on Eye Tracking Research and Applications
Proceedings of the 21st international conference on World Wide Web
Hi-index | 0.00 |
Fine-grained search interactions such as mouse cursor movements and scrolling have been shown to be valuable for modeling user attention and preferences of Web search results, in the desktop setting. However, users increasingly search the Web on touch-enabled devices such as smart phones and tablets, where they zoom and swipe instead of mousing and scrolling. In this paper, we present, to our knowledge, the first study of the utility of touch interactions on a mobile devices for estimating Web search result relevance -- which can in turn be used for search result ranking and evaluation. In particular, we explore a variety of touch interaction signals as implicit relevance feedback, based on a user study of 26 users and hundreds of unique Web search queries, result clicks, and page examinations. Our experimental results show that touch interactions provide more effective implicit feedback compared to only the time spent visiting a document, resulting in substantially higher correlation of the estimated document relevance with the explicit relevance judgments.