Title :
MDL-based segmentation of multi-attribute sequences
Author_Institution :
IBM Zurich Res. Lab., Zurich, Switzerland
fDate :
June 29 2011-July 1 2011
Abstract :
Many real-life multi-attribute sequences (multi-sequences) have a segmental structure, with segments of differing structures of attribute dependencies, that reflect an evolving nature of the dependencies over time and space. We propose a new approach for discovering a segmental structure of such evolving dependencies in probabilistic terms as a sequence of Dynamic Bayesian Networks (DBN). We use the Minimum Description Length (MDL) Principle to partition the multi-sequence into non-overlapping and homogeneous segments by fitting an optimal sequence of DBNs to the multi-sequence. In experiments, conducted on daily rainfall data we showed the applicability of the method for discovering interesting spatio-temporal evolving dependencies between rainfall occurrences in south-western Australia.
Keywords :
Bayes methods; data mining; rain; MDL-based segmentation; dynamic Bayesian networks; minimum description length; multi-attribute sequences; probabilistic terms; rainfall data; segmental structure; Artificial neural networks; Australia; Bayesian methods; Complexity theory; Markov processes; Meteorology; Probabilistic logic;
Conference_Titel :
Spatial Data Mining and Geographical Knowledge Services (ICSDM), 2011 IEEE International Conference on
Conference_Location :
Fuzhou
Print_ISBN :
978-1-4244-8352-5
DOI :
10.1109/ICSDM.2011.5969014
