Recognizing Textual Entailment Using Probabilistic Inference

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 1620–1625, Lisbon, Portugal, 17-21 September 2015. c⃝2015 Association for Computational Linguistics. Recognizing Textual Entailment Using Probabilistic Inference Lei Sha, Sujian Li, Tingsong Jiang, Baobao Chang, Zhifang Sui Key Laboratory of Computational Linguistics, Ministry of Education School of Electronics Engineering and Computer Science, Peking University Collaborative Innovation Center for Language Ability, Xuzhou 221009 China {shalei, lisujian, tingsong, chbb, szf}@pku.edu.cn Abstract Recognizing Text Entailment (RTE) plays an important role in NLP applications in- cluding question answering, information retrieval, etc. In recent work, some re- search explore “deep” expressions such as discourse commitments or strict logic for representing the text. However, these ex- pressions suffer from the limitation of in- ference inconvenience or translation loss. To overcome the limitations, in this paper, we propose to use the predicate-argument structures to represent the discourse com- mitments extracted from text. At the same time, with the help of the YAGO knowl- edge, we borrow the distant supervision technique to mine the implicit facts from the text. We also construct a probabilistic network for all the facts and conduct infer- ence to judge the confidence of each fact for RTE. The experimental results show that our proposed method achieves a com- petitive result compared to the previous work. 1 Introduction For the natural language, a common phenomenon is that there exist a lot of ways to express the same or similar meaning. To discover such dif- ferent expressions, the Recognising Textual En- tailment (RTE) task is proposed to judge whether the meaning of one text (denoted as H) can be in- ferred (entailed) from the other one (T)(Dagan et al., 2006). For many natural language processing applications like question answering, information retrieval which involve the diversity of natural lan- guage, recognising textual entailments is a critical step. PASCAL Recognizing Textual Entailment (RTE) Challenges (Dagan et al., 2006) have witnessed a variety of excellent systems which intend to recognize the textual entailment in- stances. These systems mainly employ “shallow” techniques, including heuristics, term overlap, syntactic dependencies(Vanderwende et al., 2006; Jijkoun and de Rijke, 2005; Malakasiotis and Androutsopoulos, 2007; Haghighi et al., 2005). As Hickl (2008) stated, the shallow approaches do not work well for long sentences for the missing of underlying information which needs to be mined from the surface level expression. Recently, some deep techniques are developed to mine the facts latent in the text. Hickl (2008) proposed the concept of discourse commitments which can be seen as the set of propositions in- ferred from the text, and used a series of syntax- level and semantic-level rules to extract the com- mitments from the T-H pairs. Then the RTE task is reduced to the identification of the commitments from T which are most likely to support the infer- ence of the commitments from H. From the work of Hickl (2008), we can see that a deep under- standing of text is critical to the RTE performance and discourse commitments can serve a good …

Content truncated to 500 words