An Efficient Closed Frequent Item Sets Mining Algorithm-For Mining Closed Frequent Item Sets from Data Streams

A data stream is a continuous sequence of data elements generated from a specified source. Mining frequent item sets in dynamic databases and data streams encounters some challenges that make the mining task harder than static databases. Many research works were developed in the frequent itemset mining, but these methods have the familiar problem of memory usage and processing time. Because, in data streams data elements are arrive at a rapid rate. The incoming data is unbounded and probably infinite. Due to high speed and large amount of incoming data, frequent item set mining algorithm must require a limited memory and processing time. To reduce this drawback in the existing method, a new algorithm is proposed in this paper. Here, a new algorithm is named as CFIM is developed for mining closed frequent item sets from the data streams based on their utility and consistency. During the closed frequent item sets mining, a hash table is maintained to check whether the given item set is closed or not. The computation of closed frequent item sets from the data stream will minimize the memory usage and processing time. Thus our proposed technique performance is analyzed by using the synthetic data set and compared with the exiting mining techniques.