意图识别与槽填充是语义提取的常用方法,其存在如下两个问题: 依赖训练数据,需要标注大量的数据用于训练模型;可迁移性差,训练得到的模型难以复用。针对上述问题,该文对于不同的语义提取场景,提出了四种不同的语义模型。同时,该文研究了汉语语法的特点,基于依存句法,提出了不同的语义提取算法,解决了模型难以复用的问题。该方法对数据集的要求较低,节省了成本。最后,设计了语义提取实验,验证了在样本规模小且分布不均匀的数据集下,语义提取算法相比于部分中文文本分类算法有更高的准确率。该文提出的模型和算法具有一般性,对于文本分类、人机对话等不同语义提取场景具有较强的指导意义。
Abstract
Intention recognition and slot filling are common tasks in semantic extraction, which rely on large amount of annotated training data. This paper proposes four different semantic models for different semantic extraction scenarios based on dependency syntax. This method requires less data set and saves cost. Experiments are designed to verify that the proposed methods achieve higher accuracy than some Chinese text classification algorithms in the case of small sample size and uneven distribution of data sets.
关键词
意图识别 /
依存句法 /
语义模型 /
语义提取
{{custom_keyword}} /
Key words
intention recognition /
dependency syntax /
semantic model /
semantic extraction
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 刘群, 詹卫东, 常宝宝, 等. 一个汉英机器翻译系统的计算模型与语言模型[C]//智能计算机接口与应用进展: 第二届中国计算机智能接口与智能应用学术会议论文集. 北京: 电子工业出版社, 1997: 253-258.
[2] FINK S R. Aspects of a pedagogical grammar based on case grammar and valence theory[M]. Max Niemeyer Verlag, 2011.
[3] 徐凯.现代汉语配价语法研究综述[J].今古文创,2021, 1(31): 108-109.
[4] HARRIS M. Some problems for a case grammar of Latin and early Romance[J]. Journal of Linguistics, 1975, 11(2): 183-194.
[5] BACKLEY P. Introduction to element theory[M]. Edinburgh University Press, 2011.
[6] NIVER J. Dependency parsing[J]. Language and Linguistics Compass, 2010, 4(3): 138-152.
[7] CHE W, LI Z, LIU T. LTP: A Chinese language technology platform[C]//Proceedings of the Coling: Demonstrations. 2010: 13-16.
[8] LIU T, MA J, LI S. Building a dependency treebank for improving Chinese parser[J]. Journal of Chinese Language and Computing, 2006, 16(4): 207-224
[9] BLOOMFIELD L. Linguistics and reading[J]. Language Learning, 1955, 5(3-4): 94-107.
[10] 朱德熙. 现代汉语形容词研究[J]. 语言研究, 1956, 1(1): 1-37.
[11] 邵敬敏, 周芍. 语义特征的界定与提取方法[J]. 外语教学与研究(外国语文双月刊), 2005, 37(1): 21-28.
[12] 荀恩东, 饶高琦, 肖晓悦, 等. 大数据背景下 BCC 语料库的研制[J]. 语料库语言学, 2016, 3(1): 93-118.
[13] CHEN Y. Convolutional neural network for sentence classification[D]. University of Waterloo, 2015.
[14] LIU P, QIU X, HUANG X. Recurrent neural network for text classification with multi-task learning[C]//Proceedings of the 25th International Joint Conference on Artificial Intelligence, 2016: 2873-2879.
[15] ZHOU P, SHI W, TIAN J, et al. Attention based bidirectional long short-term memory networks for relation classification[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, 2016: 207-212.
[16] LAI S, XU L, LIU K, et al. Recurrent convolutional neural networks for text classification[C]//Proceedings of the 29th AAAI Conference on Artificial Intelligence, 2015: 2267-2273.
[17] JOULIN A, GRAVE , BOJANOWSKI P, et al. Bag of tricks for efficient text classification[C]//Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics, 2017: 427-431.
[18] JOHNSON R, ZHANG T. Deep pyramid convolutional neural networks for text categorization[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, 2017: 562-570.
[19] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems, 2017: 5998-6008.
[20] PULVERMLLER F. Neurobiological mechanisms for semantic feature extraction and conceptual flexibility[J]. Topics in Cognitive Science, 2018, 10(3): 590-620.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(2624 KB)
