基于直觉模糊集的Tri-Training改进算法

彭雅琴; 宫宁生

Chinese

您当前的位置：

首页 >

文章列表页 >

基于直觉模糊集的Tri-Training改进算法

更新时间：2026-09-04

- 基于直觉模糊集的Tri-Training改进算法
- The Improvement of Tri-Training Algorithm Based on Intuitionistic Fuzzy Sets
- 微电子学与计算机 2016年33卷第3期页码：135-139
- 作者机构：
  
  1. 三江学院机构计算机科学与工程学院,江苏,南京,210012
  2. 南京工业大学机构电子与信息工程学院,江苏,南京,210009
- 作者简介：
- 基金信息：
  
  国家自然科学基金 61170322
- DOI：
  中图分类号：
- 纸质出版：2016
- 稿件说明：
移动端阅览
彭雅琴, 宫宁生. 基于直觉模糊集的Tri-Training改进算法[J]. 微电子学与计算机, 2016, 33(3): 135-139, 143.

PENG Ya-qin, GONG Ning-sheng. The Improvement of Tri-Training Algorithm Based on Intuitionistic Fuzzy Sets[J]. 2016, 33(3): 135-139.
彭雅琴, 宫宁生. 基于直觉模糊集的Tri-Training改进算法[J]. 微电子学与计算机, 2016, 33(3): 135-139, 143. DOI：

PENG Ya-qin, GONG Ning-sheng. The Improvement of Tri-Training Algorithm Based on Intuitionistic Fuzzy Sets[J]. 2016, 33(3): 135-139. DOI：

摘要

Tri-Training算法是半监督算法中的一种

其初始分类器性能受有标记样本影响较大

当样本数目不足时

分类器性能相对较弱

会直接影响后续迭代.为此提出IFS-Tri-Training(Tri-Training based on intuitionistic fuzzy sets)算法

引入SOM算法构建直觉模糊集

使得分类器在多因素下综合判别无标记样本

提高无标记样本的使用率

从而在迭代中扩展有标记样本集.在多个UCI数据上进行实验

结果数据表明

分类器的性能得到提高

学习无标记样本过程是影响分类器的关键点.

Abstract

Tri-Training algorithm belongs to semi-supervised algorithm

the initial classifier performance of the algorithm is influenced by the labeled samples. When the number of the labeled samples is insufficient

the classifier performance is relatively weak

which will affect the iteration later. Therefore

an improvement algorithm called IFS-Tri-Training(Tri-Training based on intuitionistic fuzzy sets) is proposed

the SOM algorithm is applied in the paper to build intuitionistic fuzzy sets

so the classifier can measure the unlabeled samples by more factors

which makes the efficiency of the unlabeled samples is increased and labeled sample set is expanded. The experimental result on UCI datasets shows that the classifier performance is improved and the method of studying unlabeled samples is important.

关键词

Keywords

references

Kohonen T. The Self-organizing Maps[J]. Proceedings of the IEEE, 1990, 78(9): 1464-1480. DOI: 1464

Zhou ZH, Li M. Tri-training: exploiting unlabeled data using three classifiers[J]. IEEE Trans on Knowledge and Data Engineering, 2005, 17(11): 1529-1541. DOI: 1529

BelkinM, Niyogi P, Sindhwani V. A geometric framework for learning from labeled and unlabeled examples[J]. Journal of machine learning reseach, 2006, 7(11): 2399-2434. DOI: 2399

Atanassov K T. My personal view on intuitionistic fuzzy sets theory[J]. Studies in Fuzziness and Soft Computing, 2008, 14(220): 23-43. DOI: 23

Joachims T. Transductive inference for text classification using support vector machines[C]//Proc of the 16thInt'l Conf on Machine Learning (ICML'99). San Fransisco: Morgan Kaufmann Publishers, 1999: 200-209.

Nigam K, McCallum A K, Thrun S, et al. Text classification from labeled and unlabeled documents using EM[J]. Machine learning, 2000, 39(2-3): 103-134. DOI: 103

Atanassov K T. Intuitionstic fuzzy sets[J]. Fuzzy Sets and Systems, 1986, 20(1): 87-96. DOI: 87

Cambridge OlivierC, Bernhard S, ALEXANDER Z. Semi-supervised learning[M]. Cambridge: MIT Press, 2006.

邓超, 郭茂祖. 基于Tri-Training和数据剪辑的半监督聚类算法[J]. 软件学报, 2008, 19(3): 663-673. DOI: 663

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

一种基于特征上下文理解与切片分析的肺结核检测算法

基于监控指标关联性的微服务异常检测

AlN/GaN混合缓冲层双异质结p-GaN HEMT直流特性仿真

基于多尺度融合与小目标增强的航拍车辆检测算法

注意力机制与改进密集卷积引导的信息隐藏模型