Site visiti n Helsinki on 12 June 2003

Professor Hannu Toivonen and Director Martti Mäntylä presented their project Context Recognition by User Situation Data Analysis (CONTEXT) at the Basic Research Unit of HIIT on 12 June 2003.

The aim of the project is to analyze user situation data and context data and apply the results in proactive adaptivity. A goal is to develop user-centred models of context-awareness. The research approach is multidisciplinary and includes qualitative user studies, data analysis, development of suitable algorithms and empirical testing of a prototype. The example application area is presence services and instant mobile messaging.

Current topics at BRU that are being studied are analysis of cell-based location data for presence services and discovery of personally important places or bases from location data. The group intends to use clustering of context data to discover typical contexts. A problem to solve is how the application should name the typical contexts of a user so that he or she may understand what they are. Another problem is to explain or name the context of a user so that other users understand what it is. The user of the context-aware application may be satisfied with rather implicit information about the context; other users that are trying to contact him or her need more explicit information about his or her context.

The group at ARU is involved in user studies. Among other things, they have made user observations of mobile communication of high school students and tried to detect how the willingness to communicate depends on the user's context. In order to achieve acceptability of a context-aware messaging system, the algorithms used should be simple and predictable for the user.

The project intends to build a context data logger that is able to gather context data from several different sources. Possible collected data, in addition to the above mentioned location, may include date, time, calendar information, communication with the phone, ambient sound, as well as other sensor measurements such as light, distance and temperature. The logger will also allow the user to name typical places, e.g., Office or Home. A context server may also be considered in the project.

More information
For more information, please contact the Coordinator of the project, Professor Hannu Toivonen at the University of Helsinki, see the project's web pages http://www.hiit.fi/fuego/context/ or http://www.cs.helsinki.fi/group/context/ or the web page at this site.

The research consortium consists of the following units

• University of Helsinki, Department of Computer Science and Helsinki Institute for Information Technology HIIT, Basic Research Unit (BRU): Professor Hannu Toivonen, researchers Kari Laasonen, Mika Raento
• Helsinki Institute for Information Technology HIIT Advanced Research Unit (ARU): Director Martti Mäntylä, Dr. Anu Kankainen, researchers Antti Oulasvirta, Matti Rantanen, Sauli Tiitta
  

Greger Lindén
Programme Coordinator
Greger.Linden@cs.helsinki.fi