Research Journal of Applied Sciences
ISSN: Online 1993-6079ISSN: Print 1815-932x
Abstract
A classical multilevel document classification system is vital in many contexts. This study explores the development of a multilevel document classification system based on Support Vector Machine (SVM) using ontology. SVMs have advantage over conventional statistical learning algorithms with features such as high generalization performance, prevention of over fitting, less computational complexity, high accuracy and robustness whereas the support of domain ontology further sharpens the classification by providing accurate required results. In this research SVM is used for implementing high level classification of the document and multi-level classification of the document is provided using Ontology. The comparison graph shows that the developed system based on ontology outperforms the existing system.
How to cite this article:
V. Uma and G. Aghila . SVM Based Multilevel Classifier Using Ontology.
DOI: https://doi.org/10.36478/rjasci.2007.431.434
URL: https://www.makhillpublications.co/view-article/1815-932x/rjasci.2007.431.434