Lexical-semantic SLVM for XML Document Classification

Structured link vector model (SLVM) and its improved version depend on statistical term measures to implement XML document representation. As a result, they ignore the lexical semantics of terms and its mutual information, leading to text classification errors. This paper proposed a XML document representation method, WordNet-based lexical-semantic SLVM, to solve the problem. Using WordNet, this method constructed a data structure for characterizing lexical semantic contents of XML document, and adjusted EM modeling to disambiguate word stems. Then, synset matrix of lexical semantic contents was built in the lexical-semantic feature space for XML document representation, and lexical semantic relations were marked on it to construct the feature matrix in lexical-semantic SLVM. On categorized dataset of Wikipedia XML, using NWKNN classification algorithm, the experimental results show that the feature matrix of our method performs F1 measure better than original SLVM and frequent sub-tree SLVM based on TF-IDF.