Research projects
Algorithms
Trust for All (Trust4All)
Period: 1/2005-12/2006
Researchers: Patrik Floréen, Michael Przybilski, Topi Musto
Funding: Nokia Research Center , subcontracting
The EUREKA/ITEA project Trust4All studies the concept of trust in software architectures. The focus is on context-aware systems, and especially the trust and safety features of such systems. The research is based on the middleware for embedded systems that was developed in the Robocop project and enhanced in the Space4U project.
In 2006, a model for the secure access requirements of (software) components was developed. This access control mechanism has been implemented and tested and the new approach will be integrated into the Trust4All middleware in 2007. To show the functionality of the new method, a prototype that will be utilising the new middleware has been developed. Trust4All is a EUREKA/ITEA collaboration between several European universities and companies. The research carried out at HIIT/BRU is subcontracted from Nokia Research Center .
Mobile Life (MobiLife)
Period: 9/2004-12/2006
Researchers: Patrik Floréen, Petteri Nurmi, Jukka Suomela, Fredrik Boström, Marja Hassinen, Joonas Kukkonen, Eemil Lagerspetz, Mika Karlstedt
Funding: EU IST
The objective of the MobiLife project was to bring user-centred advances in mobile applications and services within the reach of users in their everyday life. The MobiLife project, with 22 partners in nine countries, was coordinated by Nokia. It was part of a larger group of EU projects called Wireless World Initiative (WWI). The group at HIIT / BRU focused on the context awareness. It studied new context-reasoning methods, as well as developed context-aware applications.
HIIT/BRU headed the task that studied context management. MobiLife concluded at the end of year 2006.
During 2006, the open-source software BeTelGeuse (see http://www.cs.helsinki.fi/group/acs/betelgeuse/ ) for data collection was developed, and the work on a general reasoning component for producing recommendations [ 1] continued. The group also participated in developing the Context Watcher application (see http://portals.telin.nl/contextwatcher/ ). In addition, clustering algorithms for spatial data and a simulator for presenting context data were developed.
[1] Petteri Nurmi, Alfons Salden, Sian Lun Lau, Jukka Suomela, Michael Sutterer, Jean Millerat, Miquel Martin, Eemil Lagerspetz, and Remco Poortinga. A system for context-dependent user modeling. Proc. OTM Federated Workshops ( Montpellier , France , October-November 2006), Lecture Notes in Computer Science 4278. Springer-Verlag , Berlin , Germany , 2006, 1894-1903.
Semantic Interpreter Widened Experience (Stepwise)
Period: 9/2006-8/2007
Researchers: Patrik Floréen, Petteri Nurmi
Funding: Nokia Research Center , subcontracting
A subcontract project for Nokia Research Center , Stepwise is aimed at improving users' experiences of context-aware serviceswith the help of
statistical inference and simulation . The project started in September 2006.
Content-Based Retrieval and Analysis of Harmony and other Music Structures (C-BRAHMS)
Period: 8/2005-7/2010
Researchers: Kjell Lemström, Väinö Ala-Härkönen, Johan Brunberg, Niko Mikkilä, Veli Mäkinen, Anna Pienimäki, Esko Ukkonen
Funding: The Academy of Finland
The project designs and implements algorithms and data structures for analysis and content-based search of symbolically encoded music. The results of C-BRAHMS will be made available for the general public by adding them to a prototype system that will be distributed according to the conditions of the GNU GPL license. During 2006, the project developed a new prototype using a customer-server structure. It will be available on the project web page ( http://www.cs.helsinki.fi/group/cbrahms/ ) when it is completed.
Approximation and learning algorithms (ALEA)
Period: 01/2005-12/2008
Researchers: Krishnan Narayanan, Matti Kääriäinen, Jyrki Kivinen
Funding: The Academy of Finland
Approximation algorithms are methods with which to search for less-than-best solutions for computational problems. The idea is to save computational resources with problems for which it is difficult to find exact solutions. The basic method in machine learning is to create a simple hypothesis based on sample data to explain the samples. In such cases, it usually makes sense to use approximation algorithms, because it may be difficult computationally to find a simple hypothesis, and it is often unnecessary to even try to explain all the samples exactly. The samples may contain mistakes that the learning algorithm had better not copy. The project studies approximation algorithms with special attention to the needs of machine learning. One goal is to modify traditional approximation algorithm that are based on analysing worst-case scenarios, so that they can utilise the more commonly occurring easy cases better. Another approach is the development of models and methods for machine learning that bypass the notoriously difficult approximation problem, either by changing the representation of the hypothesis or by formalising the learning problem in a whole different way.