TV'03: the 3rd Workshop on Personalization in Future TV

                    in conjunction with User Modeling 2003

                             Pittsburgh, PA, USA

                               June 22nd, 2003

                          Background and motivation

There are large differences in types of users (e.g., ages, languages, ethnic
backgrounds) and ranges of viewing purposes (e.g., enjoyment, business,
education and learning, etc.) witness speciality stations in sports,
business, economics, biography/history, etc.
The diffusion of digital TV and the availability of hundreds of TV channels
are  exciting challenges  for the  design, development and  exploitation of
user  modeling, personalization  and  adaptive user  interfaces techniques.
First of  all, personalized Electronic Program  Guides (EPGs) are needed to
support the TV viewer in the selection of the interesting programs. Second,
the  presence of  Set-top  boxes running  complex programs  facilitates the
development  of user  modeling systems  that monitor the  viewer's behavior
each time  (s)he watches TV,  acquiring long-term user models.  At the same
time,  however,  there are  important  issues  to be  solved: for  example,
viewers typically  do not watch TV  alone; therefore, household models have
to be acquired and  efficient unobtrusive identification techniques have to
be developed to make the explicit identification unnecessary. Moreover, the
viewer's  interests  have to  be  acquired without  bothering her/him  with
questions and,  as such  interests may abruptly  change, the system  has to
promptly  react  to  such  changes  to  provide  suitable  recommendations.
Finally,  privacy  issues  have  to  be  addressed  to  make  such  systems

Another important  challenge for digital TV  is the development of suitable
user  interfaces  for TV  services.  For  instance, the  transition from  a
passive type  of interaction  to an active  one and the need  to use simple
remote control  devices constrains the types of  actions that the TV viewer
can  perform. Moreover,  as the TV  is located  far away from  its viewers,
there  are  special  constraints  on the  layout  of  the user  interfaces.

Finally,  increasingly  sophisticated and  interactive  games will  provide
additional challenges.  And new forms of  interactive television promise to
place the viewer in the drivers/design seat.

The previous TV workshops

The 3rd workshop  on Personalization  in Future TV  follows  TV'01, held in
association with the 8th Int. Conf. on User Modeling (Sonthofen, 2001), and
TV'02,  organized  in  conjunction with  the  2nd  Int.  Conf. on  Adaptive
Hypermedia  and  Adaptive Web Based Systems (Malaga, 2002).  Both TV events
have been organized as forums in which researchers from diverse areas  such
as machine learning,  knowledge engineering,  cognitive  sciences, adaptive
user interfaces, and business intelligence could share their experiences in
the design,  development and  exploitation of user interfaces for future TV
services.  After  the  success  of  such  workshops,  which  attracted  the
attention of academic and industrial  researchers and provided an excellent
overview of the  current international  work in the area of  digital TV, we
propose  the  third edition  of such  event, to  be  held in  Pittsburgh in
conjunction with the UM 2003 Conference.

                               Call for papers

For this workshop, we  will consider contributions devoted to user modeling
and  user-adaptive systems  in  the field  of Personalized  TV Applications
available on  the Web and on new generation TV-sets.  In order to focus the
themes addressed  during the  workshop, the submitted papers  should try to
address at least one of the following questions:

   1. Will users prefer personalized television over traditional television?

   2. What are the unknown usability problems for digital TV applications?
      What testing methods could be applied to assess the usability of TV

   3. What are the most effective algorithms for selecting shows relevant to
      viewers' personal interests?

   4. What is the most intuitive personalized interface for users to browse
      and select programs?

   5. How can privacy be managed and balanced against needs for
      personalization of massive television data?

   6. What is the best way to convey user preferences to the computer?

   7. Are stereotypes more effective than individualized models of users'
      television interests and preferences?

                               Workshop format

This  workshop is  intended to provide  a forum  in which  researchers from
diverse fields such as machine learning, knowledge engineering, psychology,
cognitive sciences,  adaptive user  interfaces, user modeling  and business
intelligence can examine the  personalization aspects of the user interface
in interactive TV.

During the workshop, the authors presenting their work should try to answer
at least  one of the previously  listed questions. Notes will be taken, in
order  to  produce a  summary  of  the contributions.  At  the  end of  the
workshop, a discussion session will be held to summarize the results of the

All contributions  will be  made available in  a Web site  before the UM'03
Conference,  so that  people  can read  them in  advance. If  possible, the
proceedings  will  also be  published  as  an informal  Annex  to the  main
Conference Proceedings.

In  order to participate to this workshop, one of the following is required:

    * present  a  contribution  to  the  workshop (long paper, short paper or
      demonstration, see below)

    * submit a short interest statement describing current work and interests
      as related to the workshop topics.

Important Dates:

    * Submission of contributions: March 14, 2003
    * Notification of acceptance: March 25, 2003
    * Submission of final contributions: May 10, 2003

Submission instructions:

The following types of submission are solicited:

    * Long paper submissions, describing substantial contributions of novel
      ongoing work. Long papers should be at most 10 pages long.

    * Short paper submissions, describing work in progress. These papers
      should be at most 5 pages long.

    * Demonstration abstracts: software demonstrations are welcome and
      should be described in a short abstract outlining the key features of
      the system to be demonstrated. The abstracts should be at most two
      pages long. Notice that the authors are expected to bring their own
      equipment for running the systems during the workshop.

All the  submissions should  be formatted according to  the guidelines used
for the papers of the main conference (see
Suitable templates (LaTex2e |  LaTex | Tex | MS Word (PC)) may be retrieved
from the LNCS Web site at the following URL:

Electronic submission of the URL address of the paper is preferred,
although e-mail submissions of the Postscript/PDF files will be accepted.
Send your submission to [log in to unmask]


Liliana Ardissono
Dipartimento di Informatica
Universita` di Torino
Corso Svizzera 185
10149 Torino
[log in to unmask]

Mark Maybury
Information Technology Division
The MITRE Corporation
MS K312
202 Burlington Road
Bedford, MA, USA
[log in to unmask]

Program Committee Members:

Liliana Ardissono (University of Torino, Italy)
Patrick Baudisch (Microsoft Research, USA)
Fabio Bellifemine (Telecom Italia Labs, Italy)
Judith Masthoff (University of Brighton, UK)
Mark Maybury (MITRE Corporation, USA)
Barry Smyth (ChangingWorlds Ltd., Ireland)
Howard Wactlar (CMU, USA)