General Terms

Abstract. The system utilizes a Recognizing Textual Entailment (RTE) approach. The system uses the question string (q_str) as a Text (T) and one answer (t_str) as a Hypothesis (H). We use two different approaches using machine learning, specifically Support Vector Machine as classifier. The results show an increment over the baselines, however enhanced is needed. The features used are unigram, bigram, and trigram overlap of lexemes and stems, Levenshtein distance, tf-idf measure, and semantic similarity using wordnet. Experimental results show that the best run of our initial system achieved a 0.21 of F-measure and 0.17 of QA-accuracy. This shows an increment of 23.53 % over the QA accuracy baseline