Title :
Discovery of context-specific Markov blankets
Author :
Klein, Assaf ; Shimony, Solomon Eyal
Author_Institution :
Dept. of Math., Ben-Gurion Univ., Beer-Sheva, Israel
Abstract :
The notion of context-specific Markov blankets (CSMB) is a refinement of Markov blankets (MB). An important property of context specific Markov-blankets is that in the worst case they need as many parameters as a model using a standard Markov blanket, but frequently considerably fewer. The result is expected to be a much improved probabilistic model of the data in practice. An algorithm for discovery of the CSMB is presented, and empirical results show that it is capable of successfully recovering context specific (\´\´local") structures.
Keywords :
Markov processes; belief networks; learning (artificial intelligence); statistical analysis; context specific structure recovery; context-specific Markov blankets; probabilistic model; Bayesian methods; Context modeling; Data mining; Knowledge representation; Learning systems; Mathematics; Probability; Random variables; Robots; Topology;
Conference_Titel :
Systems, Man and Cybernetics, 2004 IEEE International Conference on
Print_ISBN :
0-7803-8566-7
DOI :
10.1109/ICSMC.2004.1400942