Incorporating Linguistic Constraints into Keyphrase Generation

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 5224–5233 Florence, Italy, July 28 - August 2, 2019. c⃝2019 Association for Computational Linguistics 5224 Incorporating Linguistic Constraints into Keyphrase Generation Jing Zhao and Yuxiang Zhang ∗ Civil Aviation University of China, Tianjin, China jing.zhao.me@outlook.com, yxzhcn@sina.com Abstract Keyphrases, that concisely describe the high- level topics discussed in a document, are very useful for a wide range of natural language processing tasks. Though existing keyphrase generation methods have achieved remarkable performance on this task, they generate many overlapping phrases (including sub-phrases or super-phrases) of keyphrases. In this paper, we propose the parallel Seq2Seq network with the coverage attention to alleviate the overlapping phrase problem. Specifically, we integrate the linguistic constraints of keyphrases into the basic Seq2Seq network on the source side, and employ the multi-task learning framework on the target side. In addition, in order to prevent from generating overlapping phrases with cor- rect syntax, we introduce the coverage vector to keep track of the attention history and to de- cide whether the parts of source text have been covered by existing generated keyphrases. The experimental results show that our method can outperform the state-of-the-art CopyRNN on scientific datasets, and is also more effective in news domain. 1 Introduction Automatic keyphrase prediction recommends a set of representative phrases that are related to the main topics discussed in a document (Liu et al., 2009). Since keyphrases can provide a high-level topic description of a document, they are bene- ficial for a wide range of natural language pro- cessing tasks such as information extraction (Wan and Xiao, 2008), text summarization (Zhang et al., 2017) and question answering (Tang et al., 2017). However, the performance of existing methods is still far from satisfactory (Hasan and Ng, 2014). The main reason is that it is very challenging to determine whether a phrase or sets of phrases can ∗ Corresponding author accurately capture main topics that are presented in the document. Existing approaches for keyphrase prediction can be broadly divided into extraction and gener- ation methods. The conventional extraction meth- ods directly select important consecutive words or phrases from the target document as keyphrases. This means that the extracted keyphrases must appear in the target document. In comparison with extraction methods, the generation methods choose keyphrases from a predefined vocabulary regardless of whether the generated keyphrases appear in the target document. CopyRNN (Meng et al., 2017) is the first to employ the sequence-to- sequence (Seq2Seq) framework (Sutskever et al., 2014) to generate keyphrases for documents. This method is able to predict absent keyphrases that do not appear in the target document. Following the CopyRNN, a few extensions of Seq2Seq framework have been proposed to help better generate keyphrases. Through analyzing the results generated by these approaches, we find out that there are many overlapping phrases of cor- rect (author-labeled) keyphrases. For example, in experimental results of CopyRNN, the author- labeled keyphrases are “Internet” and “Distributed decision” but the predicted are “Internet held” and …

Content truncated to 500 words