Composed Association Rules for Missing Value Completion

• Main Authors: Ming-Tsung Chen, 陳銘宗
• Other Authors: none
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/7gb6vj

碩士 === 朝陽科技大學 === 資訊管理系碩士班 === 91 === The technologies of internetworking and database have expedited the distributed database systems to be implemented in the global information systems. But it cannot avoid occurring random missing values in the process of gathering data. The research on association rules can effectively establish the relationship of attributes in databases containing missing value. While this technique is used to resolve the issues on recovering the missing value, the integrity of the methodology and the measurability of the quality can already be verified in related essays. In this paper, we propose a rule recycle technique, which reuses the rules by composing them in order to retrieve more attribute association rules. The recycle composed association rules can promote the recovery rate and accuracy of missing values. Most of all, we also propose a very efficient rules mining algorithm which dramatically reduces the time complexity and space requests during rules discovery.