期刊文献+

相容关系三支聚类的治略效果评估研究

Measure Effectiveness of Action and Strategy in Three-way Clustering Based on Compatible Relationship
下载PDF
导出
摘要 三支决策是一种符合人类认知的"三分而治"模型,衡量"治"的效果需基于特定的"分"法。现有的对"治"的研究往往基于等价类进行三分。本文基于相容关系进行三支聚类,根据聚类假设对数据对象进行"治"略,提出一种测量治略效果的算法——3WC-BC。通过"治"前后的比较表明"治"略的意义和有效性。实验结果表明,三支决策中"治"略的研究具有较大意义,能够为选择合理的或最大效益的"治"略选择提供方法,与传统的离散表格索引的选择方法比较减少了因索引错误而造成的误差。 The three-way decisions are a trisecting-and-acting model that conforms to human cognition.Measuring the effectiveness of action and strategy needs to consider the specific way of secting.The existing research on action is often based on equivalence classes for three points.An algorithm of 3WC-BC is proposed for measuring effectiveness of acting in three-way clustering based on compatible relationships.The comparison between before and after action and strategy shows the significance and effectiveness of action and strategy.Experimental results show that the study of action and strategy in three-way decisions has a great significance and further provide a method for choosing reasonable or most effective actions,which reduces deviation caused by indexing error compared with traditional discrete table index method.
作者 张天麒 李英梅 ZHANG Tianqi;LI Yingmei(School of Computer Science and Information Engineering,Harbin Normal University,Harbin 150025,China)
出处 《软件工程》 2019年第2期1-4,共4页 Software Engineering
基金 黑龙江省教育厅科学研究项目(12541239) 黑龙江省自然科学基金资助(F2017021) 哈尔滨市科技创新人才研究专项基金(2016RAQXJ036)
关键词 三支决策 相容关系 三支聚类 行动 行动规则 three-way decisions compatible relation three-way cluster action rule of action
  • 相关文献

参考文献6

二级参考文献45

  • 1梁吉业,王俊红.基于概念格的规则产生集挖掘算法[J].计算机研究与发展,2004,41(8):1339-1344. 被引量:57
  • 2赵明清,蒋昌俊,陶树平.基于等价相异度矩阵的聚类[J].计算机科学,2004,31(7):183-184. 被引量:11
  • 3石志伟,刘涛,吴功宜.一种快速高效的文本分类方法[J].计算机工程与应用,2005,41(29):180-183. 被引量:15
  • 4JA1N A K, MURTY M N, FLYNN P J. Data clustering: a review [ J ]. ACM Computing Surveys, 1999,31 (3) : 264- 323.
  • 5LIU Jin-ze, ZHANG Qi, WANG Wei, et al. Clustering pairwise dissimilarity data into partially ordered sets[ C ]//Proc of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York : ACM Press, 2006:637- 642.
  • 6LIU Jin-ze, ZHANG Qi, WANG Wei, et al. PoClustering: lossless clustering of dissimilarity data[ C ]//Proc of the 7th SIAM International Conference on Data Mining. 2007:55-81.
  • 7SOCOLOVSKY E A. A dissimilarity measure for clustering high-and infinite dimensional data that satisfies the triangle inequality [ R ]. NASA LaRC Technical Library Digital Repository, 2002:1-12.
  • 8NG M K, LI M J, HUANG J Z, et al. On the impact of dissimilarity measure in k-modes clustering algorithm[ J]. IEEE Trans on Pattern analysis and Machine Intelligence, 2007,29(3):503-507.
  • 9HITCHCOCK D B, CHEN Zhi-min. Smoothing dissimilarities to cluster binary data[ J]. Computational Statistics and Data Analysis, 2008,52 (10) :4699- 4711.
  • 10VALTCHEV P, EUZENAT J. Dissimilarity Measure for collections of objects and values [ C]//Proc of Lecture Notes in Computer Sciences. Berlin : Springer, 1997:259- 272.

共引文献125

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部