Probabilistic Latent Semantic Analysis (PLSA) models have been shown to provide a better model for capturing polysemy and synonymy than Latent Semantic Analysis (LSA). However, th...
Abstract. Latent semantic indexing (LSI) is an application of numerical method called singular value decomposition (SVD), which discovers latent semantic in documents by creating c...
The work presented in this paper aims at reducing the semantic gap between low level video features and semantic video objects. The proposed method for finding associations betwee...
Convolution kernels, such as tree kernel and subsequence kernel are useful for natural language processing tasks. However, most of them ignore the semantic knowledge. In order to ...
In this paper, we introduce an integrated approach to the development of spatial hypertext. This approach brings together several theories and techniques concerning semantic struc...