基于模糊超网络的知识获取方法研究

PDF下载HTML

Title:

Fuzzy hypernetwork-based knowledge acquisition method

作者:

程麟焰^1,2, 胡峰^1,2

1. 重庆邮电大学 计算机科学与技术学院, 重庆 400065;
2. 重庆邮电大学 计算智能重庆市重点实验室, 重庆 400065

Author(s):

CHENG Linyan^1,2, HU Feng^1,2

1. College of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, China;
2. Chongqing Key Laboratory of Computational Intelligence, Chongqing University of Posts and Telecommunications, Chongqing 400

关键词:

模糊等价; 模糊集; 模糊粗糙集; 三支决策; 超网络; 知识获取方法; 分类算法

Keywords:

fuzzy equivalence; fuzzy set; fuzzy rough set; three-way decision; hypernetworks; knowledge acquisition method; classification algorithm

分类号:

TP18

DOI:

10.11992/tis.201804055

摘要:

本文结合模糊粗糙集理论与超网络的相关知识，提出了一种模糊超网络模型。与传统超网络模型的不同之处在于，模糊超网络模型采用了模糊等效关系来代替超网络中的分明等效关系，并在此基础上对超边的生成和演化进行了改进。根据样本的分布将样本集划分成3个区域，即正域、边界域和负域，不同区域的样本按照不同的方式生成超边；根据分类效果将超边集也划分成3个区域，并对不同区域的超边进行相应地替换处理。实验结果表明，在正确率、Precision、Recall等指标上，模糊超网络分类算法具有明显的优势。

Abstract:

Combining the fuzzy rough set theory with the related knowledge on hypernetworks, this paper proposes a fuzzy hypernetwork mode. In comparison with the traditional hypernetwork model, the fuzzy hypernetwork model uses the fuzzy equivalence relationship to replace the distinct equivalence relation in hypernetworks and then improves the generation and evolution of hyperedges on this basis. First, the samples are divided into three regions according to their distribution:positive, boundary, and negative regions. The samples of different regions generate hyperedges in different ways. Second, the hyperedges are also divided into three regions according to their classification results, and the corresponding replacement of hyperedges in different regions is implemented. The experimental results show that the fuzzy hypernetwork classification algorithm presents prominent advantages in terms of accuracy, precision, and recall, thus proving the validity of the classification algorithm.