Using Sequential Pattern Mining to Detect Drifts in Streaming Data

BFSPMiner is an effective and efficient batch-free algorithm for mining sequential patterns over data streams was published very recently An implementation of the algorithm is available here: As BFSPMiner has proven to be effective (see Figures 10-14 of the paper) in different domains (see Table 1 in the paper), we would like to use it for detecting drifts in the underlying stream. This can be done, for instance, by observing considerable statistical deviations in the discovered top k patterns. Additionally, for more adaptation to the streaming settings, a new method is needed to adaptively change the optimal value of the maximal pattern length.

Contact: Dr. ing. Marwan Hassani.

Leave a Reply