Critical vector learning for text categorisation

Publisher:
UTS Press
Publication Type:
Conference Proceeding
Citation:
Proceedings 4th Australasion Data Mining Conference AusDM05, 2005, pp. 27 - 36
Issue Date:
2005-01
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2005002856.pdf665.98 kB
Adobe PDF
This paper proposes a new text categorisation method based on the critical vector learning algorithm. By implementing a Bayesian treatment of a generalised linear model of identical function form to the support vector machine, the proposed approach requires signi¯cantly fewer support vectors. This leads to much reduced computational com- plexity of the prediction process, which is critical in online applications.
Please use this identifier to cite or link to this item: