Review Paper on Concept Drifting Data Stream Mining

The Data Stream in dynamic and emerging environment such ase-commerce, financial data analysis, sensor systems, socialnetworking and many more fields, that possess distribution. Theterm concepts refer to the whole distribution of the problem in acertain point in time and hence the concept drift represents achange in distribution of the problem. Data Stream thatconstantly changes with time due to some hidden concepts thatexhibit varying degree of drift, often the magnitude and thefrequency of drifting concept are not known apriori, which isvery difficult to handle, because of inadequacy of traditionaltechniques. However, with the advent of streaming data and longlife classification system, it becomes clear that training anaccurate, fast and light classifier for unpredictable, large andgrowing environment is very important and still open researchproblem.