作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2007, Vol. 33 ›› Issue (09): 219-221.

• 人工智能及识别技术 • 上一篇    下一篇

基于模拟退火遗传算法的贝叶斯分类

胡为成1,程转流1,王本年1,2   

  1. (1. 铜陵学院计算机系,铜陵 244000;2. 南京大学计算机学院,南京 240000)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-05-05 发布日期:2007-05-05

Bayesian Classification Based on Simulated Annealing Genetic Algorithms

HU Weicheng1, CHENG Zhuanliu1, WANG Bennian1,2   

  1. (1. Department of Computer, Tongling College, Tongling 244000; 2. College of Computer, Nanjing University, Nanjing 240000)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-05-05 Published:2007-05-05

摘要: 朴素贝叶斯分类器是一种简单而高效的分类器,但是其属性独立性假设限制了对实际数据的应用。文章提出一种新的算法,该算法为避免数据预处理时的属性约简对分类效果的直接影响,在训练集上通过随机属性选取生成若干属性子集,以这些子集构建相应的朴素贝叶斯分类器,采用模拟退火遗传算法进行优选。实验表明,与传统的朴素贝叶斯方法相比,该方法具有更好的性能。

关键词: 数据挖掘, 朴素贝叶斯, 模拟退火算法, 遗传算法, 属性约简, 适应度函数

Abstract: Although Naïve Bayesian classifier is a simple and highly efficient classification method, its attribute of independence assumption limits its real application. A new algorithm is introduced in this paper to avoid the direct influence of feature reduction on the performance of classification. This algorithm generates certain attribute subsets of the training sets through random attribute selection, constructs the corresponding Naïve Bayesian classifiers, and optimizes the Bayesian classifiers by using simulated annealing genetic algorithms. Experiment shows that this algorithm has better performance when compared with traditional Naïve Bayesian methods.

Key words: Data mining, Naï, ve Bayesian, Simulated annealing algorithms, Genetic algorithms, Feature reduction

中图分类号: