Efficient Mining of Weighted Frequent Patterns over Data Streams

Abstract

By considering different weights of the items, weighted frequent pattern (WFP)mining can discover more important knowledge compared to traditional frequent pattern mining. Therefore, WFP mining becomes an important research issue in data mining and knowledge discovery area. However, the existing algorithms cannot be applied for stream data mining because they require multiple database scans. Moreover, they cannot extract the recent change of knowledge in a data stream adaptively. In this paper, we propose a sliding window based novel technique WFPMDS (Weighted Frequent Pattern Mining over Data Streams) using a single scan of data stream to discover important knowledge form the recent data elements. Extensive performance analyses show that our technique is very efficient for WFP mining over data streams.