Research Issues in Data Stream Association Rule Mining

Speaker: Alan KWAN Kang Lun

Abstract

Data stream is an ordered sequence of items that arrives in continuous, unbounded and high speed fashion, which constitutes hugh amount of data and characterized by changing distribution. Association rules mining algorithms try to find the implication between itemsets from a large database of customer transactions. The traditional mining algorithms for static database is not suitable for mining the data stream characteristics due to the dynamic characteristics of streams. This paper discusses those new research issues raised for developing association rule mining techniques for stream data.