传统的关键词抽取算法不能够抽取那些没有在文本当中出现过的关键词，因此在抽取法律问题（短文本）的关键词任务上效果不佳。该文提出了一种基于强化学习的序列到序列（seq2seq）模型来从法律问题中抽取关键词。首先，编码器将给定法律问题文本的语义信息压入一个密集矢量；然后，解码器自动生成关键词。因为在关键词抽取任务中，生成的关键词的前后顺序无关紧要，所以引入强化学习来训练所提出的模型。该模型结合了强化学习在决策上的优势和序列到序列模型在长期记忆方面的优势，在真实数据集上的实验结果表明，该模型在关键词抽取任务上有较好的效果。
Traditional keyphrase extraction algorithms cannot extract keyphrases that have not appeared in the text, so they cannot effectively extract keyphrases in short legal texts. This paper presents a sequence-to-sequence (seq2seq) model based on reinforcement learning to extract keyphrases from legal questions. First, the encoder pushes the semantic information of a given legal question text into a dense vector; then, the decoder automatically generates the keyphrases. Since the order of the generated keyphrases does not matter in the keyphrase extraction task, reinforcement learning is used to train the model. This method combines the advantages of reinforcement learning for decision-making and the advantages of the sequence-to-sequence model for long-term memory. Tests on real datasets show that the model provides accurate keyphrase extraction.
[1] TURNEY P D. Learning algorithms for keyphrase extraction[J]. Information Retrieval Journal, 2002, 2(4):303-336.
[2] FRANK E, PAYNTER G W, WITTEN I H, et al. Domain-specific keyphrase extraction[C]//International Joint Conference on Artificial Intelligence. San Francisco,CA:Morgan Kaufmann Publishers,1999, 2:668-673.
[3] LIU Z, LI P, ZHENG Y, et al. Clustering to find exemplar terms for keyphrase extraction[C]//Conference on Empirical Methods in Natural Language Processing. Stroudsburg, PA:Association for Computational Linguistics, 2009:257-266.
[4] MEDELYAN O, FRANK E, WITTEN I H. Human-competitive tagging using automatic keyphrase extraction[C]//Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing. Stroudsburg, PA:Association for Computational Linguistics, 2009:1318-1327.
[5] HASAN K S, NG V. Automatic keyphrase extraction:A survey of the state of the art[C]//Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1:Long Papers). Stroudsburg, PA:Association for Computational Linguistics, 2014, 1:1262-1273.
[6] WANG M, ZHAO B, HUANG Y. PTR:Phrase-based topical ranking for automatic keyphrase extraction in scientific publications[C]//International Conference on Neural Information Processing. Berlin:Springer, 2016:120-128.
[7] BELLAACHIA A, AL-DHELAAN M. Ne-rank:A novel graph-based keyphrase extraction in twitter[C]//Proceedings of the 2012 IEEE/WIC/ACM International Joint Conferences on Web Intelligence and Intelligent Agent Technology- Volume 1. IEEE Computer Society, NJ:IEEE, 2012:372-379.
[8] ZHANG Q, WANG Y, GONG Y, et al. Keyphrase extraction using deep recurrent neural networks on Twitter[C]//Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing. Stroudsburg, PA:Association for Computational Linguistics, 2016:836-845.
[9] CHO K, VAN MERRIENBOER B, GULCEHRE C, et al. Learning phrase representations using rnn encoder-decoder for statistical machine translation[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing. Stroudsburg, PA:Association for Computational Linguistics, 2014:1724-1734.
[10] VINYALS O, KAISER , KOO T, et al. Grammar as a foreign language[C]//Advances in Neural Information Processing Systems. Cambridge, MA:MIT Press, 2015:2773-2781.
[11] RANZATO M, CHOPRA S, AULI M, et al. Sequence level training with recurrent neural networks[J/OL].(2015-11-20). https://arxiv.org/abs/1511.06732.
[12] Mihalcea R, Tarau P. Textrank:Bringing order into text[C]//Proceedings of the 2004 Conference on Empirical Methods in Natural Language Processing. Stroudsburg, PA:Association for Computational Linguistics, 2004:404-411.
[13] MNIH V, KAVUKCUOGLU K, SILVER D, et al. Playing atari with deep reinforcement learning[J/OL]. (2013-12-19). https://arxiv.org/abs/1312.5602.
[14] SUTTON R S, MCALLESTER D A, SINGH S P, et al. Policy gradient methods for reinforcement learning with function approximation[C]//Advances in Neural Information Processing Systems. Cambridge, MA:MIT Press, 2000:1057-1063.
[15] LI J, GALLEY M, BROCKETT C, et al. A diversity-promoting objective function for neural conversation models[J/OL].(2015-10-11). https://arxiv.org/abs/1510.03055.
[16] KINGMA D P, BA J. Adam:A method for stochastic optimization[J/OL].(2014-12-22). https://arxiv.org/abs/1412.6980.
[17] CHUNG J, GULCEHRE C, CHO K H, et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[J/OL].(2014-12-11). https://arxiv.org/abs/1412.3555.
[18] MENG R, ZHAO S, HAN S, et al. Deep keyphrase generation[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics. Vancouver, Canada:Association for Computational Linguistics, 2017:582-592.
