Title :
K2GA: Heuristically Guided Evolution of Bayesian Network Structures from Data
Author_Institution :
Quantum Leap Innovations, Newark, DE
fDate :
March 1 2007-April 5 2007
Abstract :
We present K2GA, an algorithm for learning Bayesian network structures from data. K2GA uses a genetic algorithm to perform stochastic search, while employing a modified version of the K2 heuristic to score proposed networks and improve future generations. We show each component of K2GA, a combination of these components to form the basic algorithm, extensions to the algorithm for improved accuracy, and numerical results
Keywords :
belief networks; genetic algorithms; learning (artificial intelligence); search problems; stochastic processes; Bayesian network structure learning; K2 heuristic; K2GA; genetic algorithm; heuristically guided evolution; stochastic search; Bayesian methods; Computational intelligence; Data mining; Genetic algorithms; Genetic mutations; Probability distribution; Quantum computing; Search methods; Stochastic processes; Technological innovation;
Conference_Titel :
Computational Intelligence and Data Mining, 2007. CIDM 2007. IEEE Symposium on
Conference_Location :
Honolulu, HI
Print_ISBN :
1-4244-0705-2
DOI :
10.1109/CIDM.2007.368847
