作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2011, Vol. 37 ›› Issue (19): 28-31. doi: 10.3969/j.issn.1000-3428.2011.19.008

• 软件技术与数据库 • 上一篇    下一篇

一种基于新型关系矩阵的数据填补方法

金成美1,鄂 旭1,2,3,穆海军1,李 岩4   

  1. (1. 辽宁工业大学电子与信息工程学院,辽宁 锦州 121001;2. 辽宁工程技术大学资源与环境学院,辽宁 阜新 123000; 3. 辽宁工程职业技术学院,辽宁 铁岭 112000;4. 中国铁通锦州分公司,辽宁 锦州 121000)
  • 收稿日期:2011-04-20 出版日期:2011-10-05 发布日期:2011-10-05
  • 作者简介:金成美(1985-),女,硕士研究生,主研方向:知识发现,智能决策支持系统;鄂 旭,教授、博士;穆海军,硕士研究生;李 岩,工程师
  • 基金资助:
    国家自然科学基金资助项目(70971059, 70771007);辽宁省博士科研启动基金资助项目“基于数据挖掘技术的电信客户市场细分系统研究”(20091034)

Data Filling Method Based on New Relationship Matrix

JIN Cheng-mei 1, E Xu 1,2,3, MU Hai-jun 1, LI Yan 4   

  1. (1. School of Electronics & Information Engineering, Liaoning University of Technology, Jinzhou 121001, China; 2. School of Resources and Environment Engineering, Liaoning Technical University, Fuxin 123000, China; 3. Liaoning Engineering and Professional Technology Institute, Tieling 112000, China; 4. China Tietong Jinzhou Branch, Jinzhou 121000, China)
  • Received:2011-04-20 Online:2011-10-05 Published:2011-10-05

摘要: 研究不完备信息系统,分析容差关系、非对称相似关系、限制容差关系的局限性,提出一种基于新型关系矩阵的数据填补方法。新型关系矩阵完整地记录了各对象之间条件属性以及决策属性的异同情况,以此挖掘对象间的潜在联系,并进行空缺值的填补处理,填补的结果不会破坏系统的协调性。数据集测试结果验证了该方法的有效性。

关键词: 新型关系矩阵, 不完备信息表, 粗糙集, 数据填补, 冲突避免

Abstract: Under incomplete information system, there are several similarity relations, such as tolerance relation, dissymmetrical similarity relation, limited tolerance relation. But all have limitations respectively. In this paper, a method for completing data based on new relationship matrix is presented. The new relationship matrix records all the situations that similarities or differences fort comparing the condition attributes and the decision attributes between objects. On basis of it, mines the potential links between objects, and completes the missing data. Results will not undermine the system’s coordination. Experimental results indicate the method is effective.

Key words: new relationship matrix, incomplete information table, rough set, data filling, collision avoidance

中图分类号: