Abstract

The goal of the project is to build probabilistic and neural computing based models that learn from the actions of people to model their intentions and expectations, and use the models for disambiguating the user's vague commands and anticipate their expectations. The actions and interests are monitored by measuring eye fixations and movements that exhibit both voluntary and involuntary signs of both the cognitive state of the user and his intentions. The models will be implemented and incorporated into a software tool offering a platform for applications in a prototype area, proactive information retrieval.

Keywords: proactive information retrieval, probabilistic models, machine learning, user modeling, eye movement tracing

Web page: http://www.cis.hut.fi/projects/mi/prima/

Site visits: 5 May 2003, 22 June 2004

Challenge: http://www.cis.hut.fi/eyechallenge2005/

Press release in English:  13 April 2005 The computer serves its user better by tracking eye gaze

Press release in Finnish: 13.4.2005 Tietokone palvelee paremmin käyttäjäänsä katsetta lukemalla

Consortium

• Academy Research Fellow Samuel Kaski, Helsinki University of Technology
• Academy Research Fellow Petri Myllymäki, Helsinki Institute for Information Technology
• Team leader Ilpo Kojo, Helsinki School of Economics

Contact information and coordinator

Academy Research Fellow Samuel Kaski, Helsinki University of Technology