作为人类语言的最小语义单位,义原已被成功应用于许多自然语言处理任务。人工构造和更新义原知识库成本较大,因此义原预测被用来辅助义原标注。该文探索了利用定义文本为词语自动预测义原的方法。词语的各个义原通常都与定义文本中的不同词语的语义有相关关系,这种现象被称为局部语义相关性。与之对应,该文提出了义原相关池化(SCorP)模型,该模型能够利用局部语义相关性来预测义原。在HowNet上的评测结果表明,SCorP取得了当前最好的义原预测性能。大量的定量分析进一步证明了SCorP模型能够正确地学习义原与定义文本之间的局部语义相关性。
Abstract
Sememes, defined as the minimum semantic units of human languages in linguistics, have been proven useful in many NLP tasks. Since manual construction and update of sememe knowledge bases (KBs) are costly, the task of automatic sememe prediction has been used to assist sememe annotation. In this paper, we explore the method of applying dictionary definitions to predicting sememes for unannotated words. We find that sememes of each word are usually semantically related to different words in its dictionary definition, and we name this matching relationship local semantic correspondence. Accordingly, we propose a Sememe Correspondence Pooling (SCorP) model which is able to capture this kind of matching to predict sememes. Evaluated on HowNet, our model is revealed with state-of-the-art performance, capable of properly learning local semantic correspondence between sememes and words in dictionary definitions.
关键词
义原预测 /
HowNet /
语义相关性
{{custom_keyword}} /
Key words
sememe prediction /
HowNet /
semantic relevance
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Bloomfield L. A set of postulates for the science of language[J]. International Journal of American Linguistics, 1926, 15(4):195-202.
[2] Dong Z, Dong Q. HowNet - A hybrid language and knowledge resource[C]//Proceedings of the 2003 International Conference on Natural Language Processing and Knowledge Engineering, IEEE, 2003.
[3] 刘群, 李素建. 基于《知网》的词汇语义相似度计算[J]. 中文计算语言学, 2002, 7(2): 59-76.
[4] Zhang Y, Gong L, Wang Y. Chinese word sense disambiguation using HowNet[C]//Proceedings of International Conference on Natural Computation 2005. Springer, Berlin, Heidelberg, 2005: 925-932.
[5] Duan X, Zhao J, Xu B. Word sense disambiguation through sememe labeling[C]//Proceedings of the 20th International Joint Conference on Arinfical in Telligence, 2007: 1594-1599.
[6] 朱嫣岚, 闵锦, 周雅倩等. 基于 HowNet 的词汇语义倾向计算[J]. 中文信息学报, 2006, 20(1): 16-22.
[7] Xianghua F, Guo L, Yanyan G, et al. Multi-aspect sentiment analysis for Chinese online social reviews based on topic modeling and HowNet lexicon[J]. Knowledge-Based Systems, 2013, 37: 186-195.
[8] 党蕾, 张蕾. 一种基于知网的中文句子情感倾向判别方法[J]. 计算机应用研究, 2010, 27(4):1370-1372.
[9] Gu Y, Yan J, Zhu H, et al. Language modeling with sparse product of sememe experts[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, 2018: 4642-4651.
[10] Niu Y, Xie R, Liu Z, et al. Improved word representation learning with sememes[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics. 2017(1): 2049-2058.
[11] Zeng X, Yang C, Tu C, et al. Chinese LIWC lexicon expansion via hierarchical classification of word embeddings with sememe attention[C]//Proceedings of the 32nd AAAI Conference on Artificial Intelligence, 2018.
[12] Xie R, Yuan X, Liu Z, et al. Lexical sememe prediction via word embeddings and matrix factorization[C]//Proceedings of the 26th International Joint Conference on Artifical Ineelligence, 2017: 4200-4206.
[13] Jin H, Zhu H, Liu Z, et al. Incorporating Chinese characters of words for lexical sememe prediction[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, 2018: 2439-2449.
[14] Li W, Ren X, Dai D, et al. Sememe prediction: Learning semantic knowledge from unstructured textual Wiki descriptions[J]. arXiv preprint arXiv:1808. 05437, 2018.
[15] Pennebaker J W, Francis M E, Booth R J. Linguistic inquiry and word count: LIWC 2001[J]. Mahway: Lawrence Erlbaum Associates, 2001, 71: 2001.
[16] Qi F, Lin Y, Sun M, et al. Cross-lingual Lexical sememe prediction[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, 2018: 358-368.
[17] Luo F, Liu T, Xia Q, et al. Incorporating glosses into neural word sense disambiguation[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, 2018: 2473-2482.
[18] Xie R, Liu Z, Jia J, et al. Representation learning of knowledge graphs with entity descriptions[C]//Proceedingsof the 30th AAAI Conference on Artificial Intelligence, 2016.
[19] Zhong H, Zhang J, Wang Z, et al. Aligning knowledge and text embeddings by entity descriptions[C]//Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, 2015: 267-272.
[20] Long T, Bengio E, Lowe R, et al. World knowledge for reading comprehension: Rare entity prediction with hierarchical LSTMs using external descriptions[C]//Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, 2017: 825-834.
[21] Hill F, Cho K, Korhonen A, et al. Learning to understand phrases by embedding the dictionary[J]. Transactions of the Association for Computational Linguistics, 2016(4): 17-30.
[22] Bosc T, Vincent P. Auto-encoding dictionary definitions into consistent word embeddings[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, 2018: 1522-1532.
[23] Thorat S, Choudhari V. Implementing a reverse dictionary, based on word definitions, using a Node-Graph Architecture[C]//Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, 2016: 2797-2806.
[24] Tissier J, Gravier C, Habrard A. Dict2vec: Learning word embeddings using lexical dictionaries[C]//Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, 2017: 254-263.
[25] Schuster M, Paliwal K K. Bidirectional recurrent neural networks[J]. IEEE Transactions on Signal Processing, 1997, 45(11): 2673-2681.
[26] Sennrich R, Haddow B, Birch A. Neural machine translation of rare words with subword units[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, 2016: 1715-1725.
[27] Li Z, Sun M. Punctuation as implicit annotations for Chinese word segmentation[J]. Computational Linguistics, 2009, 35(4): 505-512.
[28] Mikolov T, Sutskever I, Chen K, et al. Distributed representations of words and phrases and their compositionality[C]//Proceedings of Advances in Neural Information Processing Systems,2013:3111-3119.
[29] Miller G A. WordNet: A lexical database for English[J]. Communications of the ACM, 1995, 38(11): 39-41.
[30] Pennington J, Socher R, Manning C. Glove: Global vectors for word representation[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2014: 1532-1543.
[31] Kingma D P, Ba J. Adam: A method for stochastic optimization[J]. arXiv preprint arXiv:1412. 6980, 2014.
[32] Bahdanau Dzmitry, Kyunghyun Cho, Yoshua Bengio. Neural machine translation by jointly learning to align and translate[J]. arXiv preprint arXiv:1409.0473,2014.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家自然科学基金(61661146007)
{{custom_fund}}
PDF(1867 KB)
