Change Detection on Streams
Change detection and explanation on streams
A data stream is a (potentially infinite) sequence of data items x1, x2,…. As opposed to traditional data analysis, it is not assumed that the data items are generated independently from the same probability distribution. Thus change detection is an important part of data stream mining. It consists of two tasks: determining when there is a change in the characteristics of the data stream (preferably as quickly as possible) and explaining what is the nature of the change.
The nature of the data stream model means that it may be infeasible to store all of the data or to make several passes over it. For this reason, change detection algorithms should satisfy the following desiderata: the memory requirements should be constant or increase logarithmically, and the algorithm should require only one pass over the data.
There has been a lot of work on detecting change in time series data afterall of the...
- 1.Agarwal D, Barman D, Gunopulos D, Korn F, Srivastava D, Young N. Efficient and effective explanation of change in hierarchical summaries. In: Proceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2007. p.~6–15.Google Scholar
- 2.Babcock B, Babu S, Datar M, Motwani R, Wisdom J. Models and issues in data stream systems. In: Proceedings of the 21st ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems; 2002. p. 1–16.Google Scholar
- 4.Basseville M, Nikiforov IV. Detection of abrupt changes: theory and application. Englewood Cliffs, NJ: Prentice-Hall; 1993.Google Scholar
- 6.Chahrabarti S, Sarawagi S, Dom B. Mining surprising patterns using temporal description length. In: Proceedings of the 24th International Conference on Very Large Data Bases; 1998. p. 606–17.Google Scholar
- 7.Chawathe SS, Abiteboul S, Widom J. Representing and querying changes in semi-structured data. In: Proceedings of the 14th International Conference on Data Engineering; 1998. p. 4–13.Google Scholar
- 9.Dasu T, Krishnan S, Venkatasubramanian S, Yi K. An information-theoretic approach to detecting changes in multi-dimensional data streams. In: Proceedings of the 38th Symposium on the Interface of Statistics, Computing Science, and Applications; 2006.Google Scholar
- 12.Hulten G, Spencer L, Domingos P. Mining time-changing data streams. In: Proceedings of the 7th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2001. p. 97–106.Google Scholar
- 14.Kleinberg JM Bursty and hierarchical structure in streams. In: Proceedings of the 8th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2002. p. 91–101.Google Scholar
- 15.Vovk V, Nouretdinov I, Gammerman A. Testing exchangeability on-line. In: Proceedings of the 20th International Conference on Machine Learning; 2003. p. 768–75.Google Scholar