In this paper, we consider the problem of frequent elements over data stream seeks the set of items whose frequency exceeds σN for a given threshold parameter σ. We refer to this model as the sliding window model. We also use a user specified error parameter, e, to control the accuracy of the mining result. We also propose an FIA (Frequent Itemsets mining based on an Approximate counting) algorithm based on the Chernoff bound with a guarantee of the output quality and also a bound on the memory usage. The proposed algorithm show that runs significantly faster and consumes less memory than do existing algorithms for mining approximate frequent itemsets.
展开▼
