A multi-objective multi-modal optimisation approach for mining stable spatio-temporal patterns
Michele Sebag, Nicolas Tarrisson, Olivier Teytaud, Julien Lefevre and Sylvain Baillet
In: IJCAI 2005, 01-05 August 2005, Edimburg, UK.
This paper, motivated by functional brain imaging applications, is
interested in the discovery of stable spatio-temporal patterns.
This problem is formalized as a multi-objective multi-modal optimization problem: on one hand, the target patterns must show a good stability in a wide spatio-temporal region (antagonistic objectives); on the other hand, experts are interested in
finding all such patterns (global and local optima).
The proposed algorithm, termed 4D-miner, is empirically validated on artificial and real-world datasets; it shows good performances and scalability, detecting target spatio-temporal patterns within minutes from 400+ Mo datasets.