古代中文诗歌的巅峰——中文格律诗,包括律诗和绝句,是中国古典诗词的奇葩。该文从已有的古今名诗中自动学习作诗知识,实现了一个中文格律诗的自动生成系统。该系统接收用户选择的表达其思路的若干个关键词作为输入,首先,利用相关词汇数据库和语言模型,实现了根据用户选定的关键词自动生成诗歌的第一句。其次,我们独创性地将格律诗的上下句关系映射为源语言到目标语言的翻译关系,设计了一个基于短语的统计机器翻译模型,从而把诗歌的第N-1句作为输入用以生成第N句。并提供了一个用户交互式的系统,使得用户可以在每一步都选择一个最佳诗句。最后,我们还精心设计了一套翔实的格律诗评测标准,并通过单句实验和全诗实验证明,该方法是诗歌产生的一个较好的方法。
Abstract
Automatic poetry generation is considered difficult. In this paper, we propose a novel statistical approach for automatic generation of traditional Chinese metrical poetry from a few user-supplied keywords. A template-based model is used to automatically generate the first sentence of the poem. A phrase-based statistical machine translation model then generates additional sentences one-by-one. With our interactive model, the user can select the best sentence from the system’s N-best output at each step. The approach has been evaluated on the generation of quatrains of 5- and 7-character lines. The evaluation metrics for single lines as well as for the whole generated poem suggest that this method is very promising.
Key wordsartificial intelligence; machine translation; statistical machine translation; poem generation; poem evaluation
关键词
人工智能 /
机器翻译 /
统计机器翻译 /
诗歌生成 /
绝句评测
{{custom_keyword}} /
Key words
artificial intelligence /
machine translation /
statistical machine translation /
poem generation /
poem evaluation
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Long Jiang, Ming Zhou. Generating Chinese Couplets using a Statistical MT Approach[C]//The 22nd International Conference on Computational Linguistics, Manchester, England, August 2008.
[2] Philipp Koehn, Franz Josef Och, and Daniel Marcu. Statistical Phrase-Based Translation[C]//HLT/NAACL 2003.
[3] Franz Josef Och. Minimum Error Rate Training for Statistical Machine Translation[C]//ACL 2003: Proc. of the 41st Annual Meeting of the Association for Computational Linguistics, Japan, Sapporo, July 2003.
[4] Franz Josef Och, Hermann Ney. Discriminative Training and Maximum Entropy Models for Statistical Machine Translation[C]//ACL 2002: Proc. of the 40th Annual Meeting of the Association for Computational Linguistics, Philadelphia, PA, July 2002: 295-302.
[5] 罗凤珠,李元萍,曹伟政.〈中国古代诗词格律自动检索与教学系统〉[J].中文信息学报,1999,13(1): 35-42
[6] 李良炎,何中市,易勇.〈基于词连接的中文诗词风格评价技术〉[J].中文信息学报,2005.19(6):98-104.
[7] 苏劲松 , 周昌乐 , 李翼鸿.基于统计抽词和格律的全宋词切分语料库建立[J].中文信息学报,2007.21(2):52-57.
[8] Charles O. Hartman. Virtual Muse: Experiments in Computer Poetry[M]. Wesleyan University Press, 1996.
[9] Naoko Tosa, Hideto Obara and Michihiko Minoh. Hitch Haiku: An Interactive Supporting System for Composing Haiku Poem[C]//ICEC 2008: 209-216.
[10] H. Manurung, G. Ritchie and H. Thompson. Towards a computational model of poetry generation[C]//Proc. of the AISB-00 Symposium on Creative and Cultural Aspects of AI, 2001.
[11] Franz Josef Och, Nicola Ueffing, Hermann Ney. An Efficient A* Search Algorithm for Statistical Machine Translation[C]//Data-Driven Machine Translation Workshop, Toulouse, France, July 2001:55-62.
[12] Philipp Koehn. Pharaoh: a beam search decoder for phrase-based statistical machine translation models[C]//Proceedings of the Sixth Conference of the Association for Machine Translation in the Americas, 2004, pp.115-124.
[13] Andreas Stolcke. SRILM—An Extensible Language Modeling Toolkit[C]//Proc. of Intl. Conf. on Spoken Language Processing, 2002, 2: 901-904.
[14] Kishore Papineni, Salim Roukos, Todd Ward, Wei-Jing Zhu. BLEU: a Methor for automatic evaluation of machine translation[C]//Proc. of the 40th Meeting of the Association for Computational Linguistics, 2002.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金资助项目(60553001);国家自然科学基金重大资助项目(2007CB807900,2007CB807901)
{{custom_fund}}
PDF(970 KB)
