Text classification using Semantic Information and Graph Kernels

Abstract

The most common approach to the text classification problem is to use a bag-of-words representation of documents to find the classification target function. Linguistic structures such as morphology, syntax and semantic are completely neglected in the learning process. This paper uses another document representation that, while including its context independent sentence meaning, is able to be used by a structured kernel function, namely the direct product kernel. The semantic information is obtained using the Discourse Representation Theory and similarity function between documents represented by graphs is defined.