| 摘 要: 朴素贝叶斯算法是数据挖掘领域最简单的分类算法之一。为了让朴素贝叶斯能够灵活地处理连续型数 据，分类过程就需要对数据进行离散化处理。而使用模糊数学理论来解决离散化问题是一个不错的选择。因此本文考虑 将这两种方法结合，同时在去模糊化过程中引用了一种新型去模糊化方法(“内心法”)，从而生成一种新的模糊贝叶斯 混合模型。并通过一个企业评价实例简单地验证了模糊贝叶斯算法在应对连续性数据时具有良好、可靠的分类效果。 |
| 关键词: 朴素贝叶斯；模糊数学；三角模糊数；去模糊化 |
|
中图分类号: TP391
文献标识码: A
|
|
| The Application of the Improved Decision-Making Method Based on Fuzzy Bayes in the Enterprise Evaluation |
|
FENG Sijie,GUAN Jianhe1,2
|
1.( 1.China Aviation Optical-Electrical Technology Co., Ltd., Luoyang 471000, China;2.
2.China University of Geosciences(Beijing), Beijing 100083, China)
|
| Abstract: The Naive Bayes algorithm is a simple and lucid classification way in the field of data mining.When meeting with continuous data,the algorithm usually needs to make discretization in its classifying process.Luckily,the application of relevant theories about fuzzy mathematics is a good choice to solve the discretization problem.Thus,this study decides to make a combination of the Naive Bayesian algorithm and fuzzy mathematics to generate a hybrid model and,in the meanwhile,introduces a new defuzzification method (named as The incenter of area) in the classification process.Through an application case of enterprise evaluation,the fuzzy Bayesian hybrid algorithm has been proved to be effective and reliable in the process of classification for continuous data. |
| Keywords: Naive Bayes;fuzzy math;triangular fuzzy number;defuzzification |
