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
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
CLC Number:
JIN Cheng-Mei, E Xu, MU Hai-Jun, LI Yan. Data Filling Method Based on New Relationship Matrix[J]. Computer Engineering, 2011, 37(19): 28-31.
金成美, 鄂旭, 穆海军, 李岩. 一种基于新型关系矩阵的数据填补方法[J]. 计算机工程, 2011, 37(19): 28-31.