集句诗是中国古典诗歌的一种特殊体裁。是从前人的诗篇中选取已有诗句,再将其巧妙组合形成一首新诗,是一种艺术的再创造形式。集句诗的生成要求集辑而成的诗不仅合辙押韵, 且有完整的内容、连贯的上下文和新颖的主旨意境,对创作者的知识储备和诗词鉴赏能力有极高的要求。该文基于计算机的海量存储和快速检索能力,以及神经网络模型对文本语义较强的表示和理解能力,提出一种新颖的集句诗自动生成模型。该模型以数十万首古诗作为基础,利用循环神经网络(RNN)自动学习古诗句的语义表示,并设计了多种方法自动计算两句诗句的上下文关联性。根据用户输入的首句,模型能够自动计算选取上下文语义最相关连贯的诗句进行集辑,从而形成一首完整的集句诗。自动评测和人工评测的实验结果都表明,该文模型能够生成质量较好的集句诗,远远超过基线模型的效果。
Abstract
Jiju poetry is a special kind of Chinese classical poetry in which each line is selected from existing poems respectively. As a form of art recreation, the reformed poem should not only obey the structural and phonological constraints, but also have an original theme, integrated content, and coherence. In this paper, we propose a novel automatic Jiju poetry generation model based on neural network. We apply Recurrent Neural Network (RNN) to learn the vector representation of each poetry line, then we investigate different methods to measure the context coherence of two lines. Both automatic and human evaluation results show that our model can generate high-quality Jiju poems, outperforming the baseline models significantly.
关键词
关键词神经网络 /
中国古典诗歌 /
自动诗歌生成
{{custom_keyword}} /
Key words
neural network /
Chinese classical poetry /
automatic poetry generation
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 何晶,周明,蒋龙.基于统计的汉语格律诗生成研究[J].中文信息学报,2010,24(2): 96-104.
[2] Zhang X X, Mirella Lapata. Chinese poetry generation with recurrent neural networks[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing, 2014: 670-680.
[3] 周昌乐,游维,丁晓君.一种宋词自动生成的遗传算法及其机器实现[J].软件学报,2010,21(3): 427-437.
[4] Wang Q, Luo T, Wang D, Xing C. Chinese song iambics generation with neural attention-based model[C]//Proceedings of the 25th International Joint Conference on Artificial Intelligence, New York, USA, 2016: 2943-2949.
[5] Ghazvininejad M, Shi X, Jay Priyadarshi, and Kevin Knight. Hafez: an interactive poetry generation system[C]//Proceedings of ACL 2017, System Demonstrations, Association for Computational Linguistics, 2017: 43-48.
[6] Jack Hopkins J, Kiela D. Automatically generating rhythmic verse with neural networks[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, Association for Computational Linguistics, 2017: 168-178.
[7] Yi X, Li R, Sun M. Generating Chinese plassical poems with RNN encoder-decoder[C]//Proceedings of the 16th Chinese Computational Linguistics, Nanjing, China, 2017: 211-223.
[8] Bahdanau D, KyungHyun Cho K H, Bengio Y. Neural machine translation by jointly learning to align and translate[C]//Proceedings of the 2015 International Conference on Learning Representations, San Diego, CA. 2015.
[9] Gervás P. An Expert System for the Composition of Formal Spanish Poetry[J].Knowledge-Based Systerms. 2001,14(3/4): 181-188.
[10] Wu X, Tosa N, Nakatsu R. New hitch haiku: An interactive renku poem composition supporting tool applied for sightseeing navigation system[C]//Proceedings of the 8th International Conference on Entertainment Computing, Paris, France, 2009: 191-196.
[11] Robert P Levy. A computational model of poetic creativity with neural network as measure of adaptive fitness[C]//Proceedings of the ICCBR-01 Workshop on Creative Systems, 2001.
[12] Hisar Maruli Manurung H M. An evolutionary algorithm approach to poetry generation[D]. Edinburgh: University of Edinburgh,2003.
[13] Yan R, Jiang H, Lapata M, et al. I, poet: Automatic Chinese poetry composition through a generative summarization framework under constrained optimization[C]//Proceedings of the 23rd International Joint Conference on Artificial Intelligence, Beijing, China, 2013: 2197-2203.
[14] He J, Zhou M, Jiang L. Generating chinese classical poems with statistical machine translation models[C]//Proceedings of the 26th AAAI Conference on Artificial Intelligence, Toronto, Canada. 2012: 1650-1656.
[15] Yan R. i,poet: automatic poetry composition through recurrent neural networks with iterative polishing schema[C]//Proceedings of the 25th International Joint Conference on Artificial Intelligence, New York, USA, 2016: 2238-2244.
[16] Wang Z, He W, Wu H, et al. Chinese poetry generation with planning based neural network[C]//Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, Osaka, Japan, 2016: 1051-1060.
[17] Zhang J, Feng Y, Wang D, et al. Flexible and creative chinese poetry generation using neural memory[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics. Association for Computational Linguistics, 2017: 1364-1373.
[18] Yi X, Sun M, Li R, et al. Chinese poetry generation with a working memory model[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence, Stockholm, Sweden, 2018: 4553-4559.
[19] LIU, Zheng H, et al. Entity-duet neural ranking: understanding the role of knowledge graph semantics in neural information retrieval[J]. arXiv preprint arXiv: 1805.07591, 2018.
[20] Schuster M, Paliwal K K. Bidirectional recurrent neural networks[J]. IEEE Transactions on Signal Processing, 1997, 45(11): 2673-2681.
[21] Sepp Hochreiter S, Schmidhuber J A. Long short-term memory[J]. Neural computation, 1997, 9(8): 1735-1780.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家社会科学基金(13&ZD190)
{{custom_fund}}
PDF(1758 KB)
