Recognizing Textual Entailment with synthetic analysis based on SVM and feature value control

Recognizing Textual Entailment, as one of the branches of Nature Language Processing, has been widely adopted in Human Computer Interaction and Question Answering System. RTE problem is trying to build an intelligent system which can analyze the content of an input text (T), and then raises a hypothesis (H) inferred from that. My self-design RTE system, which is called SNRTE, combines lexical, syntax, and semantic 3 levels of analysis, under the support of NLP tools including Stemmer, Tokenize, Parser, POS Tag, Name Finder, WordNet2.1, and Support Vector Machine, etc. All these modules fetch useful information elements in the target text to define 49 feature values to train the system to make judgments by SVM. The training data is token from RTE official contest including 1600 pairs of tests and hypothesizes P(T,H). The average correct judgment rate is 67.5%, far above the average system correctness in RTE1 contest (55.12%) and better than the 2nd system (60.6%).