This project extended a previous MQP, the Curious Browser, to discover the effectiveness of additional implicit indicators of user interest. We researched possible indicators, and then conducted a small study to find which new indicators should be added. Two examples were the quantization of horizontal and vertical mouse movement. Amongst other results we showed that the number of clicks on a page provided a strong prediction of the explicit rating while the amount of mouse wheel use did not.
Download
Related publications:
Mark Claypool, Phong Le, Makoto Waseda and David Brown, Implicit Interest Indicators, In Proceedings of ACM Intelligent User Interfaces Conference (IUI), Santa Fe, New Mexico, January 14-17, 2001.
Mark Claypool, David Brown, Phong Le, Makoto Waseda. Inferring User Interest, IEEE Internet Computing, November/December 2001. /~claypool/papers/iui/