Eztraction of Meta-Rules for RFID Mining Based on a Concept Hierarchy
Recently, Radio Frequency Identification (RFID) technology is being deployed for several applications, including supply-chain optimization, business process automation, asset tracking, and problem traceability applications. The problem with RFID data is that its degree increases according to time and location, thus, resulting in an enormous volume of data duplication. Therefore it is difficult to extract useful knowledge hidden in data using existing association rule mining techniques, or analyze data using statistical techniques or queries. The mining method proposed in this paper improves mining of association rules by defining meta-rules and only uses user specified parameters, viz. minsup and minconf. Also, it reduces the complexity of rule generation by using a pre-defined meta-rule to limit the generation of association rules to the level of interest to the consumer, instead of the entire concept hierarchy. As a result, rule generation time is reduced and there is a significant increase in query speed, due to filtering of data.
Younghee Kim Umo Kim
Department of Computer Engineering, Sungkyunkwan University, Suwon, Korea Department of Computer Engineering, Sungky-unkwan University, Suwon, Korea
国际会议
武汉
英文
232-237
2008-12-19（万方平台首次上网日期，不代表论文的发表时间）