Call for Papers

NIPS 2005 Workshop on Machine Learning for Implicit Feedback and User Modeling

Inferring Relevance from Eye Movements Challenge 2005

Core event of the PASCAL EU Network of Excellence

Contents

Background

The workshop is intended to gather together machine learning researchers interested in a new promising and challenging research area. Half of the papers will introduce best-performing solutions of the Challenge of inferring intent of users based on eye movement signals. The rest will broaden the scope towards other implicit feedback signals such as click streams, and towards more general problems of user modeling.

The tasks require advanced signal processing and feature extraction, and dynamic machine learning models. Several branches of machine learning are applicable, and we expect eye movement data to become a new challenging testbench for machine learning algorithms.

We have made eye movement data available in the form of a Challenge where the objective is to predict from eye movement data whether a reader finds a text relevant. The scientific goals of the Challenge are to advance machine learning methodology, to find the best eye movement features, and to learn of the psychology underlying eye movements in search tasks. The challenge has been organized for the PASCAL EU Network of Excellence, but participation is open to all.

Submission

Papers are sought in two categories:

  1. papers based on Challenge submissions, and
  2. regular papers.

Both will be refereed, and may be accepted as oral or poster presentations. Shorter position papers are possible.

A non-comprehensive list of relevant phrases: Adaptive user interfaces, Eye movements, Human-computer interfaces, Implicit feedback, Machine learning methods, Modeling of user intent, Personalization, Proactive systems, Statistical modeling, User modeling, (connection to) Vision research

The papers will be made available in the Web, and either a special issue of a journal or an edited book is being planned of selected papers.

Length: 2-6 pages. Follow the NIPS format (http://www.nips.cc/Conferences/current/CFP/CallForPapers.php). Submissions violating these guidelines will not be considered.

Please submit the paper by email (inips2005@mail.cis.hut.fi) in postscript or PDF format by 21 October 2005. Please contact us in advance if your file is unusually large (over 10 MB). Include the email address of the corresponding author in the email. We will verify the submission by email by 24 October 2005. The acceptance notice will be emailed to the corresponding author by mid November, with instructions on how to submit the camera-ready paper by 21 November 2005.

Pascal Invited Talk

Greg Edwards, founder and CTO of Eyetools Inc.

Brief description from www.eyetools.com:

Eyetools was born in 2000 out of the Stanford University Advanced Eye Interpretation Project. After seeing the business value of eyetracking resulting from the Stanford-Poynter Project, a collaborative study between the Poynter Institute and Stanford University's Department of Communications around the viewing of online news sites, founder Greg Edwards spun out Eyetools. Since then, Eyetools' pioneering work in inferring mental state from eye movements and visualizing eyetracking data has led to several key patents in the area, and has enabled eyetracking to be put into use more easily by an ever expanding number of companies and people. Eyetools' roots in Human-Computer Interaction began in 1995.

Venue

NIPS 2005 Workshops, Whistler, December 10. More details at http://www.nips.cc/Conferences/2005/

Deadlines

September 30
Submissions to Challenge 1 (easier one)
October 14
Submissions to Challenge 2 (more difficult)
October 21
Deadline for papers
By mid November:
Notification of acceptance
November 21
Camera ready papers
December 10
Workshop

Note: Challenge deadlines have been extended because of the late notice for the NIPS workshop.

Program Committee

Samy Bengio, IDIAP
Helene Hembrooke, Cornell
Thorsten Joachims, Cornell
Samuel Kaski, Helsinki University of Technology
Ilpo Kojo, Helsinki School of Economics
Petri Myllymäki, University of Helsinki
Kai Puolamäki, Helsinki University of Technology
Kari-Jouko Räihä, University of Tampere
John Shawe-Taylor, University of Southampton

Organizers

Samuel Kaski and Kai Puolamäki
Helsinki University of Technology
Laboratory of Computer and Information Science
P.O. Box 5400
FI-02015 TKK, Finland
Email: inips2005@mail.cis.hut.fi

NIPS 20005 Workshop on Machine Learning for Implicit Feedback and User
Modeling           PASCAL - Pattern Analysis, Statistical Modelling and Computational
Learning

This CFP as a text file, PDF file.

URL of this page: http://www.cis.hut.fi/inips2005/