张兴,王海荣,李明亮,王栋

• 1:北方民族大学计算机科学与工程学院

摘要(Abstract)：

为解决医疗问答系统精准度不高和深层语义匹配不准确的问题,提出了融合TF-IDF、BERT和DSSM的问答模型(TIBD-QA)。在妈妈网中抓取儿童病护理的相关数据,构建了儿童病护理问答数据集ChildQA,使用DSSM算法解决人工特征转换效率低的问题;利用BERT中的多头注意力机制使模型可以关注不同方面的信息,使得到的上下文信息更加准确。对比实验结果表明,所提方法在ChildQA数据集和WebQA公开数据集上的精确率分别达到83.6%和84.3%,且在构建的儿童病问答系统上取得较好效果。

关键词(KeyWords)： 问答系统;BERT模型;知识图谱;DSSM

Abstract:

Keywords:

基金项目(Foundation): 宁夏自然科学基金项目(2020AAC03218);; 宁夏产教融合示范专业项目(2018SFZY14);; 北方民族大学教育教学改革重点项目(2019ZDJY01)

作者(Author): 张兴,王海荣,李明亮,王栋

DOI: 10.13705/j.issn.1671-6841.2021317

参考文献(References)：

• [1] WESTON J,BORDES A,COPRA S,et al.Towards AI-complete question answering:a set of prerequisite toy tasks[EB/OL].[2021-02-18].https://www.researchgate.net/publication/272522139.
• [2] HII P C,CHUNG W Y.A comprehensive ubiquitous healthcare solution on an androidTM mobile device[J].Sensors,2011,11(7):6799-6815.
• [3] LUO K Q,LIN F L,LUO X S,et al.Knowledge base question answering via encoding of complex query graphs[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing.Stroudsburg:Association for Computational Linguistics,2018:2185-2194.
• [4] ZHU S G,CHENG X,SU S.Knowledge-based question answering by tree-to-sequence learning[J].Neurocomputing,2020,372:64-72.
• [5] BORDES A,CHOPRA S,WESTON J.Question answering with subgraph embeddings[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing.Stroudsburg:Association for Computational Linguistics,2014:615-620.
• [6] DONG L,WEI F R,ZHOU M,et al.Question answering over freebase with multi-column convolutional neural networks[C]//Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing.Stroudsburg:Association for Computational Linguistics,2015:260-269.
• [7] XU K,LAI Y,FENG Y,et al.Enhancing key-value memory neural networks for knowledge based question answering[C]//Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics.Stroudsburg:Association for Computational Linguistics,2019:2937-2947.
• [8] CHEN Y,WU L F,ZAKI M J.Bidirectional attentive memory networks for question answering over knowledge bases[C]//Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics.Stroudsburg:Association for Computational Linguistics,2019:2913-2923.
• [9] OCASIO W.Attention to attention[J].Organization science,2011,22(5):1286-1296.
• [10] 张芳芳，马敬东，王小贤，等.国外医学领域自动问答系统研究现状及启示[J].医学信息学杂志，2017,38(3):2-6,25.ZHANG F F,MA J D,WANG X X,et al.Research status and implications of foreign medical automatic question answering system[J].Journal of medical informatics,2017,38(3):2-6,25.
• [11] 李超.智能疾病导诊及医疗问答方法研究与应用[D].大连：大连理工大学，2016.LI C.Research and application on intelligent disease guidance and medical question answering method[D].Dalian:Dalian University of Technology,2016.
• [12] 颜昕.基于自然语言处理的医疗健康问答系统[J].通讯世界，2018(6):255-256.YAN X.Medical health question answering system based on natural language processing[J].Telecom world,2018(6):255-256.
• [13] 毕铭文，左敏，张青川.基于LSTM-SPA的医学领域问答技术研究[J].山东工业技术，2019(1):247,239.BI M W,ZUO M,ZHANG Q C.Research on question answering technology in medical field based on LSTM-SPA [J].Shandong industrial technology,2019(1):247,239.
• [14] WAN S X,LAN Y Y,GUO J F,et al.A deep architecture for semantic matching with multiple positional sentence representations[EB/OL].[2021-02-18].https://www.researchgate.net/publication/285271115.
• [15] 李苗苗，邢凯，张利萍，等.基于图计算和知识图谱的疾病辅助诊别研究[J].电子技术，2018,47(9):8-12.LI M M,XING K,ZHANG L P,et al.Research on disease auxiliary diagnosis based on graph computing and knowledge graph[J].Electronic technology,2018,47(9):8-12.
• [16] 昝红英，窦华溢，贾玉祥，等.基于多来源文本的中文医学知识图谱的构建[J].郑州大学学报(理学版),2020,52(2):45-51.ZAN H Y,DOU H Y,JIA Y X,et al.Construction of Chinese medical knowledge graph based on multi-source corpus[J].Journal of Zhengzhou university (natural science edition),2020,52(2):45-51.
• [17] 马晨浩.基于甲状腺知识图谱的自动问答系统的设计与实现[J].智能计算机与应用，2018,8(3):102-107.MA C H.Design and implementation of automatic question answering system based on thyroid knowledge map[J].Intelligent computer and applications,2018,8(3):102-107.