Encyclopedia of Database Systems

Living Edition
| Editors: Ling Liu, M. Tamer Özsu

Classification in Streams

Living reference work entry
DOI: https://doi.org/10.1007/978-1-4899-7993-3_52-2



The classification problem is a well defined problem in the data mining domain, in which a training data set is supplied, which contains several feature attributes, and a special attribute known as the class attribute. The class attribute is specified in the training data, which is used to model the relationship between the feature attributes and the class attribute. This model is used in order to predict the unknown class label value for the test instance.

A data stream is defined as a large volume of continuously incoming data. The classification problem has traditionally been defined on a static training or test data set, but in the stream scenario, either the training or test data may be in the form of a stream.

Historical Background

The problem of classification has been studied so widely in the classification literature, that a single source for the problem cannot be identified. Most likely, the problem was...


Data Stream Incoming Data Association Rule Mining Concept Drift Credit Card Fraud 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
This is a preview of subscription content, log in to check access.

Recommended Reading

  1. 1.
    Aggarwal CC, editor. Data streams: models and algorithms. Berlin/Heidelberg/New York: Springer; 2007.Google Scholar
  2. 2.
    Domingos P, Hulten G. Mining high speed data streams. In: Proceedings of the 6th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2000. p. 71–80.Google Scholar
  3. 3.
    James M. Classification algorithms. New York: Wiley; 1985.Google Scholar

Copyright information

© Springer Science+Business Media LLC 2016

Authors and Affiliations

  1. 1.IBM T. J. Watson Research CenterYorktown HeightsUSA