| 基于信息论和遗传算法的Bayesian网络弧定向方法研究 |
| |
| 引用本文： | 李小琳,何湘东,苑森淼.基于信息论和遗传算法的Bayesian网络弧定向方法研究[J].复旦学报(自然科学版),2004,43(5):717-720. |
| |
| 作者姓名： | 李小琳 何湘东 苑森淼 |
| |
| 作者单位： | 吉林大学,计算机科学与技术学院,长春,130012;吉林大学,数学研究所,长春,130012 |
| |
| 基金项目： | 国家自然科学基金资助项目(60275026)，吉林省自然科学基金资助项目(20030517 1) |
| |
| 摘 要： | Bayesian网弧定向问题是Bayesian网学习的一个重要方面．提出了一种基于信息论和遗传算法的Bayesian网弧定向算法．将信息论中鉴别信息这一概念引入Bayesian网学习中来，以鉴别信息定向后的网络为基础网，并设计相应的适应度函数和遗传算子，使该算法能够收敛到全局最优的Bayesian网结构．从而极大地减弱了单纯利用遗传算法学习对初始群体的依赖性，提高了算法的收敛速度．实验结果表明：该算法能够有效地解决Bayesian网弧定向问题．
|
| 关 键 词： | Bayesian网 信息论 鉴别信息 遗传算法 计算复杂度 |
| 文章编号： | 0427-7104(2004)05-0717-04 |
Research on Orienting Edges of Bayesian Network Based on Information Theory and Genetic Algorithms |
| |
| LI Xiao-lin,HE Xiang-dong,YUAN Sen-miao.Research on Orienting Edges of Bayesian Network Based on Information Theory and Genetic Algorithms[J].Journal of Fudan University(Natural Science),2004,43(5):717-720. |
| |
| Authors: | LI Xiao-lin HE Xiang-dong YUAN Sen-miao |
| |
| Institution: | LI Xiao-lin~1,HE Xiang-dong~2,YUAN Sen-miao~1 |
| |
| Abstract: | Orienting edges of Bayesian network is part and parcel of learning Bayesian Network. An algorithm is proposed based on information theory and genetic algorithms. Cross-entropy is introduced in learning Bayesian network. Based on the network which oriented edges with cross-entropy, fitness function and genetic operators are designed, it provides guarantee of convergence. This algorithm can weaken the dependence of initial population and increase the convergence speed. Experimental result shows that this algorithm can effectively orient edges of Bayesian network. |
| |
| Keywords: | Bayesian networks information theory cross-entropy genetic algorithm computational complexity |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
