Batch-Incremental Learning for Mining Data Streams

Holmes, Geoffrey; Kirkby, Richard Brendon; Bainbridge, David

Batch-Incremental Learning for Mining Data Streams

dc.contributor.author	Holmes, Geoffrey
dc.contributor.author	Kirkby, Richard Brendon
dc.contributor.author	Bainbridge, David
dc.date.accessioned	2009-01-08T01:41:38Z
dc.date.available	2009-01-08T01:41:38Z
dc.date.issued	2004
dc.description.abstract	The data stream model for data mining places harsh restrictions on a learning algorithm. First, a model must be induced incrementally. Second, processing time for instances must keep up with their speed of arrival. Third, a model may only use a constant amount of memory, and must be ready for prediction at any point in time. We attempt to overcome these restrictions by presenting a data stream classification algorithm where the data is split into a stream of disjoint batches. Single batches of data can be processed one after the other by any standard non-incremental learning algorithm. Our approach uses ensembles of decision trees. These tree ensembles are iteratively merged into a single interpretable model of constant maximal size. Using benchmark datasets the algorithm is evaluated for accuracy against state-of-the-art algorithms that make use of the entire dataset.	en
dc.format.mimetype	application/pdf
dc.identifier.citation	Holmes, G., Kirkby, R. & Bainbridge, D.(2004). Batch-Incremental Learning for Mining Data Streams. Working papers, University of Waikato, Department of Computer science 2004, Hamilton, New Zealand: University of Waikato.	en
dc.identifier.uri	https://hdl.handle.net/10289/1749
dc.language.iso	en
dc.publisher	University of Waikato	en_NZ
dc.subject	computer science	en
dc.subject	classification option trees	en
dc.subject	ensemble methods	en
dc.subject	data streams	en
dc.subject	Machine learning
dc.title	Batch-Incremental Learning for Mining Data Streams	en
dc.type	Working Paper	en
pubs.elements-id	53736
uow.relation.series	Department of Computer Science Working Papers	en_NZ