基于自然最近邻的样本扰动三支聚类
Perturbation three-way clustering based on natural nearest neighbors
朱金 1付玉 2管文瑞 3王平心4
作者信息
- 1. 江苏科技大学经济管理学院,江苏 镇江 212100
- 2. 南京中医药大学镇江附属医院(镇江中医院),江苏 镇江 212000
- 3. 江苏科技大学自动化学院,江苏 镇江 212100
- 4. 江苏科技大学理学院,江苏 镇江 212100
- 折叠
摘要
利用数据样本的自然最近邻信息,给出了一种基于样本扰动理论的三支聚类算法,结合自然最近邻信息生成 2 组扰动数据集,随机提取特征子集并使用K-means聚类算法获得不同的聚类结果,利用共现概率矩阵和确定函数获得样本的稳定性,根据样本稳定性阈值将样本划分为稳定区域和不稳定区域,再对 2 个区域的样本使用不同的策略获得每个类簇的核心域和边界域.实验采用 5 个公开数据集与 2 种传统的聚类算法进行对比,结果验证了所提算法的有效性.
Abstract
By using sample's natural nearest neighbors,a three-way clustering algorithm is proposed based on sample's perturbation theory.The proposed algorithm combines natural nearest neighbor information with sample's perturbation to generate two datasets.By randomly selecting parts of the sample's feature,different clustering results are obtained through K-means clustering algorithms.The stability of each sample is calculated based on the defined frequencies.The universe is divided into stable set and unstable set based on the sample's stability.Then,we use different strategies to obtain the core region and fringe region of each cluster.The testing results on five open datasets verify the effectiveness of the proposed algorithm through comparative tests with two traditional clustering methods.
关键词
三支决策/三支聚类/样本扰动/自然最近邻Key words
three-way decision/three-way clustering/sample's perturbation/natural nearest neighbor引用本文复制引用
基金项目
国家自然科学基金(62076111)
国家自然科学基金(61773012)
江苏省高等学校自然科学研究项目(15KJB110004)
出版年
2024
NETL
NSTL
万方数据