Fuzzy Meta-Association Rules

Abstract

Association rules is a useful tool to extract new information from raw data expressed in a comprehensive way for decision makers. However, in some applications raw data might not be available for several reasons. First, stream data are only temporarily available for their processing or if it is stored, only summaries or representations of the extracted knowledge are kept. Second, under some circumstances primary data cannot be disclosed due to privacy or legal restrictions. In the light of these observations we propose fuzzy meta-association rules for mining association rules over already discovered rules in a set of databases sharing common information. We compare this proposal with a previous one using crisp meta-rules showing that fuzzy meta-association rules discover interesting knowledge obtaining a more manageable set of rules for human inspection and allowing the use of fuzzy items to express additional knowledge about the original databases.