A Hybrid Clustering Algorithm for the Measurement of Subspace

doi:10.7652/xjtuxb201803019

您当前的位置：

首页 >

文章列表页 >

A Hybrid Clustering Algorithm for the Measurement of Subspace

更新时间：2025-07-09

- A Hybrid Clustering Algorithm for the Measurement of Subspace
- Vol. 52, Issue 3, Pages: 139-144+167(2018)
- 作者机构：
  
  西安交通大学机械工程学院,西安,710049
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb201803019
  CLC： TP181
- Online First：10 March 2018，
  
  Published：2018
- 稿件说明：
移动端阅览
A Hybrid Clustering Algorithm for the Measurement of Subspace[J]. 2018, 52(3): 139-144+167.
DOI：

A Hybrid Clustering Algorithm for the Measurement of Subspace[J]. 2018, 52(3): 139-144+167. DOI： 10.7652/xjtuxb201803019.

摘要

针对闵可夫斯基子空间聚类算法对特征权重分配的问题

提出了一种混合测量子空间聚类算法(iMWK-HD)

以实现调节特征权重因子和提高算法性能的目的。利用闵可夫斯基距离与余弦相结合的混合测量来分配特征权重

构造新的目标函数; 在聚类迭代过程中

采用智能K-means进行初始化来解决选择正确类数的问题; 根据新的目标函数

使用拉格朗日乘子法求解新的隶属度和特征权重更新公式

使类中心更加稳定

从而促进特征空间转换

获取数据集最优聚类结果。采用UCI数据集设计了对比实验

实验结果表明

iMWK-HD算法优于iK-means、iWK-means、iMWK-means这3个现有的聚类算法

所提算法能有效提升聚类精确度和聚类结果的稳定性。

Abstract

This study presents a hybrid clustering algorithm for the measurement of subspace. This approach is adopted to distribute the feature weight in Minkowski algorithm to adjust the feature weighting factor and improve the performance of the algorithm. The feature weight is assigned by using hybrid dissimilarity measurement of Minkowski distance and Cosine dissimilarity and a new objective function is designed. In the process of clustering iteration

the problem of selecting correct class number is solved by using intelligent K-means initialization. According to the new objective function

the Lagrange multiplier method is used to solve the new membership degree and the feature weight iteration updating formula

so that the class center is more stable and the optimal clustering result of dataset is obtained by promoting the transformation of feature space. Adopting UCI dataset to design the experiments

the results showed that compared with the other three algorithms

i.e.

iK-means

iWK-means and iMWK-means algorithms

the proposed algorithm can effectively improve the clustering accuracy and the stability of clustering results.

关键词

Keywords

references

PAN Y, WANG J, DENG Z. Alternative soft subspace clustering algorithm [J]. Journal of Information & Computational Science, 2013, 10: 3615-3624.

CHEN X, YE Y, XU X, et al. A feature group weighting method for subspace clustering of high-dimensional data [J]. Pattern Recognition, 2012, 45: 434-446.

MACQUEEN J. Some methods for classification and analysis of multivariate observations [J]. Proceedings of the 5th Berkeley Symposium on Mathematical Statistics and Probability, 1967, 1: 281-297.

HUANG Z, NG M K, RONG H, et al. Automated variable weighting in k-means type clustering [J]. IEEE Transactions on Pattern Analysis and Machine Learning, 2005, 27(5): 657-668.

RENATO C D, MIRKIN B. Minkowski metric, feature weighting and anomalous cluster initializing in K-Means clustering [J]. Pattern Recognition, 2012, 45: 1061-1075.

KIM D. Group-theoretical vector space model [J]. International Journal of Computer Mathematics, 2015, 92(8): 1536-1550.

WANG L, HAO Z, CAI R, et al. Enhanced soft subspace clustering through hybrid dissimilarity [J]. Journal of Intelligent Fuzzy Systems, 2015, 29: 1395-1405.

CHIANG M M, MIRKIN B. Intelligent choice of the number of clusters in k-means clustering: an experimental study with different cluster spreads [J]. Journal of Classification, 2010, 27(1): 1-38.

PARSONS L, HAQUE E, LIU H. Subspace clustering for high dimensional data: a review [J]. ACM Sigkdd Explorations Newsletter, 2004, 6(1): 90-105.

DENG Z, CHOI K S, CHUNG F L, et al. Enhanced soft subspace clustering integrating within-cluster and between-cluster information [J]. Pattern Recognition, 2010, 43(3): 767-781.

HUANG J Z, XU J, NG M, et al. Weighting method for feature selection in k-means [J]. Computational Methods of Feature Selection, Chapman Hall/CRC, 2007, 193-209.

RAND W M. Objective criteria for the evaluation of clustering methods [J]. Journal of the American Statistical Association, 1971, 66: 846-850.

兰进,徐亮,马永浩,等.实验研究类螺纹孔旋流冲击射流的冷却特性.2018,52(1):8-13.[doi:10.7652/xjtuxb201801 002]

韩树楠,张旻,李歆昊.高容错(2,1,m)卷积码快速盲识别方法.2017,51(12):28-34.[doi:10.7652/xjtuxb201712005]

姜洪权,王岗,高建民,等.一种适用于高维非线性特征数据的聚类算法及应用.2017,51(12):49-55.[doi:10.7652/xjtuxb201712008]

曹岩,韦婉钰,乔虎,等.采用设计结构矩阵的突发事件扩散预测及建模方法.2017,51(12):68-75.[doi:10.7652/xjtuxb 201712011]

王瑞琴,高炎,晏鑫,等.燃气透平尾缘开缝区冷却性能的非定常研究.2017,51(12):98-103.[doi:10.7652/xjtuxb201712 015]

程传奇,郝向阳,李建胜,等.融合改进A*算法和动态窗口法的全局动态路径规划.2017,51(11):137-143.[doi:10.7652/xjtuxb201711019]

黄兴旺,曾学文,郭志川,等.采用改进二进制蝙蝠算法的任务调度算法.2017,51(10):65-70.[doi:10.7652/xjtuxb2017 10011]

舒帆,屈丹,张文林,等.采用长短时记忆网络的低资源语音识别方法.2017,51(10):120-127.[doi:10.7652/xjtuxb2017 10020]

曹卫权,褚衍杰,李显.针对机器学习中残缺数据的近似补全方法.2017,51(10):142-148.[doi:10.7652/xjtuxb201710 023]

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Windage Loss and Flow Characteristics in Impeller Back Gaps of sCO2 Radial Inflow Turbines

Turbulent Kinetic Energy Dissipation Mechanism of Supercritical Carbon Dioxide Leakage Flow in a Scallop Bionic Seal Cavity

Knowledge Graph Representation Learning Driven by Semantic Discrimination and Subgraph Integration

Mechanism and Optimal Control Strategy for Hysteresis in Rapid Dehumidification Response of Indirect Thermal Systems in Electric Vehicles

Effects of Steam Addition on the Swirl Combustion and Emission Characteristics of Cracked Ammonia

Related Author

No data

Related Institution

No data

⁰