在智慧医疗中基于知识图谱的问答系统能够根据结构化的医疗知识自动回答自然语言问句,具有重要的研究意义和实际应用价值。当前的问答系统不能有效地处理包含多种意图的复杂问句,导致意图识别不全面或不正确,难以生成高质量的答案。因此,该文提出了基于语义分析和深度学习的复杂问句意图智能理解方法,首先从问句中提取医疗实体并进行依存句法分析,通过句法成分规范化将多意图复杂问句分解成若干属性类或关系类简单问句的组合,然后构建文本分类深度网络模型对每个简单问句进行意图识别,从而实现复杂问句的意图理解。为了验证该文方法的有效性和实用性,该文构建了包含6类约14万个实体的医疗知识图谱,用所提出的意图理解方法为核心开发了基于知识图谱的医疗咨询智能问答系统,根据问句意图将相应的核心实体和关系谓词转化为知识图谱检索语句,并通过检索到的相关知识生成自然语言答案。对真实医疗咨询问句测试的结果表明,该文方法可以有效地理解复杂问句的多种意图,相应的问答系统能够更全面、准确地回答与疾病、症状、药品等相关的医疗咨询问句。
Abstract
In intelligent medical service, current QA system cannot deal with the complex question with multiple intentions. This paper proposes an intelligent understanding method of complex question based on semantic analysis and deep learning. The medical entity extraction and dependency parsing are first performed on the input question. Then, the syntax standardization method is proposed to decompose the input multi-intention question into several simple questions about attribute or relation. Finally, the intent understanding of the whole sentence is accomplished by classifying each simple question with a deep neural network. To validate the effectiveness of the proposed method, this paper builds a medical KG containing about 140,000 entities of 6 typical categories. The retrieval query is generated with the corresponding core entities and the relational predicates in the question intention, and the retrieved knowledge from the KG are generated as the answer. The results on the real medical consultation questions show that the proposed method can effectively recognize the multiple intentions in the complex questions and the corresponding QA system can produce comprehensive and accurate answers.
关键词
意图理解 /
问句规范化 /
人机问答
{{custom_keyword}} /
Key words
intention understanding /
syntax standardization /
human-robot question answering
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] CAI Q,YATES A. Large-scale semantic parsing via schema matching and lexicon extension[C]//Proceedings of Annual Meeting of the Association for Computational Linguistics,2013: 423-433.
[2] BERANT J,CHOU A,FROSTIG R,et al. Semantic parsing on freebase from question-answer pairs[C]//Proceedings of Conference on Empirical Methods in Natural Language Processing,2013: 1533-1544.
[3] YAO X,VANDURME B. Information extraction over structured data: Question answering with freebase[C]//Proceedings of Annual Meeting of the Association for Computational Linguistics,2014: 956-966.
[4] DONG L,WEI F,ZHOU M,et al. Question answering over freebase with multi-column convolutional neural networks[C]//Proceedings of Annual Meeting of the Association for Computational Linguistics,2015: 260-269.
[5] BORDES A,CHOPRA S,WESTON J. Question answering with subgraph embeddings[C]//Proceedings of Conference on Empirical Methods in Natural Language Processing,2014: 615-620.
[6] ZETTLEMOYER L S,COLLINS M. Learning to map sentences to logical form: Structured classification with probabilistic categorial grammars[J]. arXiv preprint arXiv: 1207.1420,2012.
[7] 周博通,孙承杰,林磊,等. 基于LSTM的大规模知识库自动问答[J]. 北京大学学报(自然科学版),2018,54(2): 286-292.
[8] 曹明宇,李青青,杨志豪,等. 基于知识图谱的原发性肝癌知识问答系统[J]. 中文信息学报,2019,33(6): 88-93.
[9] CUI W,XIAO Y,WANG H,et al. KBQA: Learning question answering over QA corpora and knowledge bases[C]//Proceedings of the VLDB Endowment,Munich,2017: 565-576.
[10] WANG K,MING Z Y,HU X,et al. Segmentation of multi-sentence questions: Towards effective question retrieval in CQA services[C]//Proceedings of the 33rd International ACM SIGIR Conference on Research and Development in Information Retrieval,2010: 387-394.
[11] KALYANPUR A,PATWARDHAN S,BOGURAEV B,et al. Fact-based question decomposition for candidate answer re-ranking[C]//Proceedings of the 20th ACM International Conference on Information and Knowledge Management,2011: 2045-2048.
[12] 刘雄,张宇,张伟男,等. 基于依存句法分析的复合事实型问句分解方法[J]. 中文信息学报,2017,31(3): 140-146.
[13] HU S,ZOU L,ZHANG X. A state-transition framework to answer complex questions over knowledge base[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing,2018: 2098-2108.
[14] CHEN X,SHI Z,QIU X,et al. Adversarial multi-criteria learning for Chinese word segmentation[C]//Proceedings of Annual Meeting of the Association for Computational Linguistics,2017: 1193-1203.
[15] LI H,HAGIWARA M,LI Q,et al. Comparison of the impact of word segmentation on name tagging for Chinese and Japanese[C]//Proceedings of the 9th International Conference on Language Resources and Evaluation,2014: 2532-2536.
[16] 王若佳,魏思仪,王继民. BiLSTM-CRF模型在中文电子病历命名实体识别中的应用研究[J]. 文献与数据学报,2019,2: 53-66.
[17] HUANG Z,XU W,YU K. Bidirectional LSTM-CRF models for sequence tagging[J].arXiv preprint arXiv: 1508.01991,2015.
[18] BUCHHOLZ S,MARSI E. CoNLL-X shared task on multilingual dependency parsing[C]//Proceedings of the 10th Conference on Computational Natural Language Learning,2006: 149-164.
[19] ZHANG M,ZHANG Y,CHE W,et al.Character-level Chinese dependency parsing[C]//Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics,2014: 1326-1336.
[20] KIM Y.Convolutional neural networks for sentence classification[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing,2014: 1746-1751.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家重点研发计划(2018YFB1305200);国家自然科学基金(61801178);广东省重点领域研发计划(2018B010107001)
{{custom_fund}}
PDF(7172 KB)
