Recognizing Textual Entailment using Dependency Analysis and Machine Learning

This paper presents a machine learning system that uses dependency-based features and lexical features for recognizing textual entailment. The proposed system evaluates the feature values automatically. The performance of the proposed system is evaluated by conducting experiments on RTE1, RTE2 and RTE3 datasets. Further, a comparative study of the current system with other ML-based systems for RTE to check the performance of the proposed system is also presented. The dependency-based heuristics and lexical features from the current system have resulted in significant improvement in accuracy over existing state-of-art ML-based solutions for RTE.