Mining Frequent Item over Evolving Data Streams Based On Decay Function Model

Because of the fluidity and continuity of data stream, the knowledge embedded in stream data is most likely to be changed as time goes by. Thus, in most data stream applications, people are more interested in the information of the recent transactions than that of the old. A data structure DF-FI is put forward based on this foundation. An algorithm DFStream which based on decay function for mining frequent item over data streams is put forward. The DFStream algorithm can accurately mining the frequent item over the sliding window. The IBM synthesizes data generation which output customer shopping a data and the accessing record of the world cup official website data in 1998 are adopted as experiment data. The algorithm has a very high preciseness for mining frequent item, and fast ability for data processing.