基于事件语义角色的常识知识获取

PDF(2426 KB)

中文信息学报 ›› 2023, Vol. 37 ›› Issue (6) : 77-88.

知识表示与知识获取

基于事件语义角色的常识知识获取

王亚^1,2,曹存根¹,王石¹

作者信息 +

Commonsense Knowledge Acquisition Via Semantic Roles in Events

WANG Ya^1,2, CAO Cungen¹, WANG Shi¹

Author information +

History +

摘要

常识知识获取是知识获取的瓶颈问题,该文提出了利用事件语义角色进行常识知识获取的方法。首先获取独立于句子(事件)的语义角色相关的常识知识(例如,施事有意志)和语义角色在句子中涉及的常识知识(例如,施事有嘴是施事喝受事的必要条件),该文将这些常识知识看作常识知识获取模板;接着使用一个神经网络模型对大规模Web文本进行语义角色的自动标注;最后利用标注结果中的文本内容替换常识知识获取模板中的语义角色来进一步获取常识知识。实验结果表明,该方法可以获取大量高精度的常识知识。此外,该文通过组合事件谓词(简称谓词)和标注结果中语义角色对应的文本内容生成一个谓词的下位谓词短语或事件,进而自动获取上下位关系常识。

Abstract

The commonsense knowledge acquisition is known as a bottleneck issue. This paper proposes an approach to acquire commonsense knowledge via semantic roles in events. Firstly, we collect commonsense knowledge about semantic roles independent of sentences (events) (e.g. agents act of their own volition) and semantic roles involved in sentences (e.g. an agent must have a mouth to drink). This kind of commonsense knowledge is regarded as commonsense knowledge acquisition templates. We then use a neural network model to automatically label semantic roles in large-scale Web corpus. Finally, the semantic roles in a commonsense knowledge acquisition template are replaced by the text contents in the annotation results to further enrich commonsense knowledge. Experimental results show that this approach is effective for acquiring large amounts of high-quality commonsense knowledge. In addition, we can automatically obtain commonsense knowledge about isa relations by generating hyponyms (i.e. phrases or events) of an event predicate (predicate for short) by combining this predicate with text contents corresponding to semantic roles.

导出引用

王亚,曹存根,王石. 基于事件语义角色的常识知识获取. 中文信息学报. 2023, 37(6): 77-88

WANG Ya, CAO Cungen, WANG Shi. Commonsense Knowledge Acquisition Via Semantic Roles in Events. Journal of Chinese Information Processing. 2023, 37(6): 77-88

参考文献

[1] WANG H, ZHANG Y, JI Z, et al. Consensus-aware visual-semantic embedding for image-text matching[C]//Proceedings of European Conference on Computer Vision. Springer, Cham, 2020: 18-34.
[2] GARG A, TANDON N, VARDE A S. I am guessing you can't recognize this: generating adversarial images for object detection using spatial commonsense (student abstract)[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2020: 13789-13790.
[3] ZHOU H, YOUNG T, HUANG M, et al. Commonsense knowledge aware conversation generation with graph attention[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence, 2018: 4623-4629.
[4] WU S, LI Y, ZHANG D, et al. Diverse and informative dialogue generation with context-specific commonsense knowledge awareness[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 5811-5820.
[5] GHOSAL D, HAZARIKA D, ROY A, et al. KinGDOM: Knowledge-guided domain adaptation for sentiment analysis[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 3198-3210.
[6] VIJAYARAGHAVAN P, ROY D. Modeling human motives and emotions from personal narratives using external knowledge and entity tracking[C]//Proceedings of the Web Conference, 2021: 529-540.
[7] LENAT D B. CYC: A large-scale investment in knowledge infrastructure[J]. Communications of the ACM, 1995, 38(11): 33-38.
[8] SINGH P, LIN T, MUELLER E T, et al. Open mind common sense: Knowledge acquisition from the general public[C]//Proceedings of Confederated International Conferences "On the Move to Meaningful Internet Systems". Berlin, Heidelberg, 2002: 1223-1237.
[9] SAP M, LE BRAS R, ALLAWAY E, et al. Atomic: An atlas of machine commonsense for if-then reasoning[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2019: 3027-3035.
[10] SPEER R, CHIN J, HAVASI C. Conceptnet 5.5: An open multilingual graph of general knowledge[C]//Proceedings of the 31th AAAI Conference on Artificial Intelligence. San Francisco, USA, 2017: 4444-4451.
[11] VON AHN L, KEDIA M, BLUM M. Verbosity: A game for collecting common-sense facts[C]//Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 2006: 75-78.
[12] LIEBERMAN H, SMITH D, TEETERS A. Common consensus: A web-based game for collecting commonsense goals[C]//Proceedings of the ACM Workshop on Common Sense for Intelligent Interfaces, 2007.
[13] ZHANG H, KHASHABI D, SONG Y, et al. Transomcs: From linguistic graphs to commonsense knowledge[J].Proceedings of the International Joint Conference on Artificial Intelligence, 2020: 3976-3982.
[14] MALAVIYA C, BHAGAVATULA C, BOSSELUT A, et al. Commonsense knowledge base completion with structural and semantic context[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2020: 2925-2933.
[15] TANDON N, DE MELO G, SUCHANEK F, et al. Webchild: Harvesting and organizing commonsense knowledge from the web[C]//Proceedings of the 7th ACM International Conference on Web Search and Data Mining, 2014: 523-532.
[16] LI X, TAHERI A, TU L, et al. Commonsense knowledge base completion[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, 2016: 1445-1455.
[17] TANDON N, HARIMAN C, URBANI J, et al. Commonsense in parts: Mining part-whole relations from the web and image tags[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2016: 243-250.
[18] MOHAMED M, OUSSALAH M. SRL-ESA-TextSum: A text summarization approach based on semantic role labeling and explicit semantic analysis[J]. Information Processing & Management, 2019, 56(4): 1356-1372.
[19] OSMAN A H, SALIM N, BINWAHLAN M S, et al. An improved plagiarism detection scheme based on semantic role labeling[J]. Applied Soft Computing, 2012, 12(5): 1493-1502.
[20] PYATKIN V, ROIT P, MICHAEL J, et al. Asking it all: Generating contextualized questions for any semantic role[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing, 2021: 1429-1441.
[21] ZHANG J, SONG X, ZENG Y, et al. Taxonomy completion via triplet matching network[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2021: 4662-4670.
[22] CHO Y, RODRIGUEZ J D, GAO Y, et al. Leveraging WordNet paths for neural hypernym prediction[C]//Proceedings of the 28th International Conference on Computational Linguistics, 2020: 3007-3018.
[23] FILLMORE C J. The case for case[M]. Universals in Linguistic Theory, 1968.
[24] 朱晓亚. 现代汉语句模研究[M]. 北京: 北京大学出版社, 2001.
[25] 袁毓林. 基于认知的汉语计算语言学研究[M]. 北京: 北京大学出版社, 2008.
[26] 鲁川. 知识工程语言学[M]. 北京: 清华大学出版社, 2010.
[27] 刘茂福, 胡慧君. 基于认知与计算的事件语义学研究[M]. 北京: 科学出版社, 2013.
[28] DONG Z, DONG Q. HowNet: A hybrid language and knowledge resource[C]//Proceedings of the International Conference on Natural Language Processing and Knowledge Engineering, 2003: 820-824.
[29] KONDREDDI S K, TRIANTAFILLOU P, WEIKUM G. Combining information extraction and human computing for crowdsourced knowledge acquisition[C]//Proceedings of the 30th International Conference on Data Engineering, 2014: 988-999.
[30] 郑怀徳, 孟庆海. 汉语形容词用法词典[M]. 北京: 商务印书馆, 2003.
[31] COLLELL G, VAN GOOL L, MOENS M F. Acquiring common sense spatial knowledge through implicit spatial templates[C]//Proceedings of the 32th AAAI Conference on Artificial Intelligence, 2018: 6765-6772.
[32] MILLER G A. WordNet: An electronic lexical database[M]. MIT Press, 1998.
[33] SUCHANEK F M, KASNECI G, WEIKUM G. Yago: A core of semantic knowledge[C]//Proceedings of the 16th International Conference on World Wide Web, 2007: 697-706.
[34] WU W, LI H, WANG H, et al. Probase: A probabilistic taxonomy for text understanding[C]//Proceedings of the ACM SIGMOD International Conference on Management of Data, 2012: 481-492.

基金

国家重点研究与发展计划(2017YFC1700300,2017YFB1002300);国家自然科学基金(61702234)

PDF(2426 KB)

869

Accesses

Citation

Detail

段落导航

摘要
Abstract
关键词
Key words
引用本文
参考文献
基金

选择文件类型/文献管理软件名称

选择包含的内容

摘要

Abstract

关键词

Key words

引用本文

{{custom_sec.title}}

{{custom_sec.title}}

参考文献

{{custom_fnGroup.title_cn}}

脚注

基金



Issue Date
2023-08-15