Tensor Space Model for Hypertext Representation

We investigate the basics of tensor based hypertext representation and perform experiments this novel hypertext representation model. Most documents have an inherent hierarchical structure that render the desirable use of multidimensional representations such as those offered by tensor objects. We focus on the advantages of Tensor Space Model, in which documents are represented using second-order tensors. We exploit the local-structure and neighborhood recommendation encapsulated by the proposed representation. We define the distance metric on tensor space of hypertext documents, which is a generalization of distance metric defined on vector space model. Our results provide evidence that tensor based model is very efficient for clustering and classification of hypertext documents compared to traditional vector based model.