Bayesian network structure optimization method and device based on BRKGA algorithm

• Main Authors: ZHANG WEIMING, ZHU CHENG, SUN BAODAN, ZHU XIANQIANG, ZHOU YUN
• Format: Patent
• Language: Chinese; English
• Published: 22.10.2021
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention relates to a Bayesian network structure optimization method and device based on a BRKGA algorithm. The method comprises the following steps: generating a plurality of random key vectors consisting of a plurality of random numbers between [0, 1]; taking the NO TEARS algorithm as a decoder to decode the random key vector, and obtaining an initial solution of the Bayesian network structure optimization problem; calculating an adaptive value of each initial solution according to the initial solutions and a preset loss function; sorting the adaptive values to obtain an elite individual set; according to the elite individual set, carrying out crossover and mutation operation by adopting a genetic algorithm, and outputting a global approximate solution of a Bayesian network structure optimization problem, wherein the global approximate solution is a continuous value matrix; and discretizing the continuous value matrix to obtain a global approximate adjacency matrix, and obtaining an optimized Bayesian