Learning Analogy-Preserving Sentence Embeddings for Answer Selection

Proceedings of the 23rd Conference on Computational Natural Language Learning, pages 910–919 Hong Kong, China, November 3-4, 2019. c⃝2019 Association for Computational Linguistics 910 Learning Analogy-Preserving Sentence Embeddings for Answer Selection A¨ıssatou Diallo†,‡ , Markus Zopf‡ and Johannes F¨urnkranz‡ †Research Training Group AIPHES ‡Knowledge Engineering Group Department of Computer Science, Technische Universit¨at Darmstadt {diallo@aiphes, mzopf@ke, juffi@ke}.tu-darmstadt.de Abstract Answer selection aims at identifying the cor- rect answer for a given question from a set of potentially correct answers. Contrary to pre- vious works, which typically focus on the se- mantic similarity between a question and its answer, our hypothesis is that question-answer pairs are often in analogical relation to each other. Using analogical inference as our use case, we propose a framework and a neu- ral network architecture for learning dedicated sentence embeddings that preserve analogical properties in the semantic space. We evaluate the proposed method on benchmark datasets for answer selection and demonstrate that our sentence embeddings indeed capture analog- ical properties better than conventional em- beddings, and that analogy-based question an- swering outperforms a comparable similarity- based technique. 1 Introduction Answer selection is the task of identifying the correct answer to a question from a pool of can- didate answers. The standard methodology is to prefer answers that are semantically similar to the question. Often, this similarity is strengthened by bridging the lexical gap between the text pairs via learned semantic embeddings for words and sentences. The main drawback of this method is that question-answer (QA) pairs are modeled independently, and that the correspondence be- tween different pairs is not considered in these embeddings. In fact, these methods only focus on the relationship that may exist between the entities that constitutes the QA pair at hand and are thus, limited to pairwise semantic structures. Instead, we argue in this paper that questions and their correct answers often form analogical relations. For example, the question ”Who is the president of the United States?” and its answer are Figure 1: Illustration of analogy-based answer selec- tion. Given a question and its candidate answers, each pair is compared to a QA prototype pair. The candidate answer with the highest score is assumed to be the correct answer. in the same relation to each other as the question ”Who is the current chancellor of Germany?” and ”Angela Merkel”. Thus, for modelling these relations, we need to look at quadruples of textual items in the form of two question-answer pairs, and want to reinforce that they are in the same relation to each other. We expect that using analogies to identify and transfer positive relationships between QA pairs will be a better approach for tackling the task of answer selection than simply looking at the similarity between individual questions and their answers. We use sentence embeddings as the mechanism to assess the relationship between two sentences, and aim to learn a latent representation in which their analogical relation is explicitly enforced in the latent space. Analogies are defined as rela- tional similarities between two …

Content truncated to 500 words