基于最小平方和残差的高阶模糊联合聚类算法

基金项目: 国家自然科学基金资助项目71272216 60903080 博士后科学基金资助项目2012M5100480 国家科技支撑计划资助项目2009BAH42B02 2012BAH08B02中央高校基本科研业务费专项基金资助项目HEUCFZ1212 HEUCF100603

详细信息

A Minimum Sum-squared Residue for High-order Fuzzy Co-clustering Algorithm

¹College of Computer Science & Technology Harbin Engineering University Harbin 150001 China

摘要: 目前,多数高阶联合聚类算法属于硬划分方法,不考虑聚簇重叠问题。为了更有效地分析具有重叠聚簇结构的数据,提出了一种基于最小平方和残差的高阶模糊联合聚类算法(MSR-HFCC),该算法将聚类问题转化为最小化模糊平方和残差的优化问题,推导出求解优化问题的隶属度迭代更新公式,设计出聚类过程的迭代算法。实验结果表明,MSR-HFCC算法聚类效果优于目前已有的5种硬划分高阶联合聚类算法。
- 高阶异构数据 /
- 联合聚类 /
- 模糊聚类 /
- 平方和残差
Abstract: Most existing high-order co-clustering algorithms focus on hard clustering methods，which ignore the problem of overlaps in the clustering structures. In order to analyze the clustering results of data with overlapping clusters more efficiently，we developed a minimum sum-squared residue for high-order fuzzy co-clustering algorithm(MSR-HFCC).The clustering problem is formulated as the problem of minimizing fuzzy sum-squared residue.The update rules for fuzzy memberships were derived，and an iterative algorithm was designed for a co-clustering process.Finally，experimental recults show that the qualities of clustering results of MSR-HFCC are superior to five existing algorithms.
- high-order heterogeneous data /
- co-clustering /
- fuzzy clustering /
- sum-squared residue

1.	李钦，游雄，李科，汤奋，王玮琦. 局部物体块匹配的图像匹配算法. 武汉大学学报(信息科学版). 2022(03): 419-427 . 百度学术
2.	张乾，杨玉成，邵定琴，岳诗琴. 广义局部二值模式在彩色图像检索中的仿真. 计算机仿真. 2020(06): 457-461 . 百度学术