Skip to the content

SVM categorizer: a generic categorization tool using support vector machines

Kapoutsis, E, Theodoulidis, B and Saraee, M 2004, SVM categorizer: a generic categorization tool using support vector machines , in: IC-AI 2004, 21-24 June 2004, Las Vegas, USA.

PDF - Published Version
Download (320kB) | Preview


Supervised text categorisation is a significant tool considering the vast amount of structured, unstruc-tured, or semi-structured texts that are available from internal or external enterprise resources. The goal of supervised text categorisation is to assign text documents to finite pre-specified categories in order to extract and automatically organise information coming from these resources. This paper pro-poses the implementation of a generic application – SVM Categorizer using the Support Vector Ma-chines algorithm with an innovative statistical adjustment that improves its performance. The algo-rithm is able to learn from a pre-categorised document corpus and it is tested on another uncatego-rized one based on a business intelligence case study. This paper discusses the requirements, design and implementation and describes every aspect of the application that will be developed. The final output of the SVM Categorizer is evaluated using commonly accepted metrics so as to measure its per-formance and contrast it with other classification tools.

Item Type: Conference or Workshop Item (Paper)
Additional Information: ISBN: 1-932415-32-7
Themes: Media, Digital Technology and the Creative Economy
Schools: Colleges and Schools > College of Science & Technology > School of Computing, Science and Engineering > Data Mining and Pattern Recognition Research Centre
Journal or Publication Title: Proceedings of the International Conference on Machine Learning; Models, Technologies and Applications
Publisher: CSREA Press
Refereed: Yes
Depositing User: Dr Mo Saraee
Date Deposited: 02 Nov 2011 11:58
Last Modified: 20 Aug 2013 17:16

Actions (login required)

Edit record (repository staff only) Edit record (repository staff only)


Downloads per month over past year