|
|
|
     |
                                         |
Return to Categorization and supervised machine learning
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.
@inproceedings{1076081, author = {Dell Zhang and Xi Chen and Wee Sun Lee}, title = {Text classification with kernels on the multinomial manifold}, booktitle = {SIGIR '05: Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval}, year = {2005}, isbn = {1-59593-034-5}, pages = {266--273}, location = {Salvador, Brazil}, doi = {http://doi.acm.org/10.1145/1076034.1076081}, publisher = {ACM Press}, address = {New York, NY, USA}, }  ©2006 Association for Computing Machinery |
