Please use this identifier to cite or link to this item: https://doi.org/10.1145/1076034.1076081
Title: Text classification with kernels on the multinomial manifold
Authors: Zhang, D. 
Chen, X.
Lee, W.S. 
Keywords: differential geometry
kernels
machine learning
manifolds
support vector machine
text classification
Issue Date: 2005
Source: Zhang, D.,Chen, X.,Lee, W.S. (2005). Text classification with kernels on the multinomial manifold. SIGIR 2005 - Proceedings of the 28th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval : 266-273. ScholarBank@NUS Repository. https://doi.org/10.1145/1076034.1076081
Abstract: Support Vector Machines (SVMs) have been very successful in text classification. However, the intrinsic geometric structure of text data has been ignored by standard kernels commonly used in SVMs. It is natural to assume that the documents are on the multinomial manifold, which is the simplex of multinomial models furnished with the Riemannian structure induced by the Fisher information metric. We prove that the Negative Geodesic Distance (NGD) on the multinomial manifold is conditionally positive definite (cpd), thus can be used as a kernel in SVMs. Experiments show the NGD kernel on the multinomial manifold to be effective for text classification, significantly outperforming standard kernels on the ambient Euclidean space. © 2005 ACM.
Source Title: SIGIR 2005 - Proceedings of the 28th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval
URI: http://scholarbank.nus.edu.sg/handle/10635/78377
ISBN: 1595930345
DOI: 10.1145/1076034.1076081
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

47
checked on Feb 13, 2018

Page view(s)

32
checked on Feb 17, 2018

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.