Improving Binary-class Chinese Textural Entailment by monolingual machine translation technology

In this paper, we describe how we improve our system for Chinese Textual Entailment Recognition by a monolingual machine translation system. Previously, our approach is based on the standard supervised learning classification. We integrate the result of monolingual machine translation system with the other available computational linguistic resources of Chinese language processing to build the system for the natural language processing application. We observed the training corpus and list all possible features. The features include surface text, semantic and syntactical information, such as POS tagging, synonym substitution, and dependency relation. The annotated data is used in training statistical models and build the classifier for the Binary-class Chinese textual Entailment Recognition task. The experiment result shows that the monolingual machine translation technology can improve the system performance in both 10-fold cross validation and open test.