Hypertext Classification Using Tensor Space Model and Rough Set Based Ensemble Classifier
PReMI '09 Proceedings of the 3rd International Conference on Pattern Recognition and Machine Intelligence
| Hi-index | 0.00 |
We investigate the basics of tensor based hypertext representation and perform experimentsthis novel hypertext representation model. Most documents have an inherent hierarchical structure that renderthe desirable use of multidimensional representations such as those offered by tensorobjects. We focus on the advantages of Tensor Space Model, in whichdocuments are represented using second-order tensors. We exploitthe local-structure and neighborhood recommendation encapsulated by the proposed representation. We define the distance metric on tensor space of hypertext documents, which is a generalizationof distance metric defined on vector space model.Our results provide evidence that tensor based model is very efficient for clustering and classificationof hypertext documents compared to traditional vector based model.