Robust Clustering of Data Streams using Incremental Optimization

Discovering the patterns in evolving data streams is a very important and challenging task. In many applications, it is useful to detect the dierent patterns evolving over time and be able to capture them accurately (e.g. detecting the purchasing trends of customers over time on an e-commerce website). Data stream mining is challenging because of harsh constraints due to the continuous arrival of huge amounts of data that prevent unlimited storage and processing in memory, and the lack of control over the data arrival pattern. In this paper, we present a new approach to discover the evolving clusters in a data stream by incrementally updating the density of the clusters using a method based on robust statistics. Our approach shows robustness toward an unknown amount of outliers, with no assumptions about the number of clusters. Moreover, it can adapt to the evolution of the clusters in the input data.

• Type: Conference
• Year: 2010
• Where: AMW
• Authors: Basheer Hawwash, Olfa Nasraoui
• Keywords: AMW 2010, Data Stream, Data Stream Mining, Evolving, Knowledge Management