Data Stream Model-Issues, Challenges and Clustering Techniques

Applications such as satellite networks, telecommunication systems etc., are generating massive amount of continuous data streams every day. It is not possible to manage and analyze the streams with conventional database techniques and architectures. Hence, data streams require efficient database storage, retrieve, processing ability, and effective analysis methods to discover knowledge which is implicit in streams. Data stream mining which is one of the steps of Knowledge discovery process (KDD), is the extraction of useful knowledge structures from continuous chunks of data with high data rate. Since, incoming data is continuous; it is not possible to scan the record more than once. Stream mining is a hot research area unveils number of challenges unsolved by traditional systems. In this paper, we explore data streams showing how the Data stream management system (DSMS) is different from traditional Data base management system (DBMS), the issues and challenges in data stream mining, and tools for data stream mining.