A Survey on Frequent Pattern Mining Over Data Streams

Frequent pattern mining is one of the important tasks used in data mining domain. Frequent pattern mining is used to find interesting patterns from databases, such as association rules, correlations rules, sequence rules, classifier rules, and cluster rules. The main goal of the association rule is, to analyze the purchased products of a customer in a supermarket transactional data. Association rule is used to describe how frequently items are purchased together. It is mainly used in transactional data base. Data streams (12) are an ordered sequence of items that arrives in timely order. It is impossible to store the data in which item arrives. To apply data mining algorithm directly to streams instead of storing them before in a database. Real time surveillances system, telecommunication system, sensor network, financial applications, transactional data are some of the examples of the data stream systems. These types of streams produced millions or billions of updates every hour. As data stored in a database and data warehouse are processed by using some mining algorithm. Data mining (1) is defined as the process of extracting information or interesting pattern or end product from huge amount of data. In this paper, we have studied the concept of data streams and how the frequent patterns are mined from data streams. We also analyzed the existing research works in the field of frequent pattern mining data streams.