Industry knowledge base construction based on knowledge graph
-
摘要: 首先阐述了机器学习和知识图谱的相关概念以及在行业知识库建设中的应用情况和地位,然后结合典型算法介绍了机器学习常见的模型,为提高行业知识库中知识的关联性并降低冗余性,引入了行业知识图谱及其构建相关的新技术方法,进而引出了对于行业知识库构建方法的研究,结合智能知识库展示了知识图谱的创新性应用,即利用知识图谱为知识库的搜索和推荐功能提供技术支持,同时通过知识图谱对领域知识进行更加直观地展示。最后,结合行业知识库的建设工作对机器学习和知识图谱在其中的作用发挥进行了更深一步的阐述和总结。Abstract: This paper firstly expounds the related concepts of machine learning and knowledge map, and their applications in the construction of knowledge base and the positions in the industry. Then introduces the common model of machine learning combined with the typical algorithm. To increase the relevance of industry knowledge in the knowledge base and reduce the redundancy, this paper introduces a new technology related industry knowledge map and its construction method, thus led to the study of the method of building knowledge base for industry, and combined with intelligent knowledge base shows the innovative application of knowledge map, using the knowledge map to provide technical support for the search and recommendation feature of the knowledge base, at the same time through knowledge map shown more visually for the domain knowledge. Finally, this paper further describe the role of machine learning and knowledge map play in industry knowledge.
-
Key words:
- knowledge map /
- machine learning /
- deep learning /
- natural language processing /
- knowledge base
-
表 1 Bert和Ernie在测试集上的表现
模型/指标 precision recall f1 Bert 0.918 069 0.928 892 0.928 892 Ernie 0.940 827 0.949 102 0.944 946 -
[1] 张栋豪, 刘振宇, 郏维强, 等. 知识图谱在智能制造领域的研究现状及其应用前景综述[J]. 机械工程学报, 2021, 57(5): 90-113. [2] 王萌, 王昊奋, 李博涵, 等. 新一代知识图谱关键技术综述[J/OL]. 计算机研究与发展: 1-18[2022-03-16]. http://kns.cnki.net/kcms/detail/11.1777.TP.20220301.1217.002.html [3] 王昊奋, 漆桂林. 知识图谱: 方法、实践与应用[M]. 北京: 电子工业出版社, 2019. [4] 廖开际, 黄琼影, 席运江. 在线医疗社区问答文本的知识图谱构建研究[J]. 情报科学, 2021, 39(3): 51-59, 75. [5] 钱玲飞, 崔晓蕾. 基于数据增强的领域知识图谱构建方法研究[J]. 现代情报, 2022, 42(3): 31-39. doi: 10.3969/j.issn.1008-0821.2022.03.004 [6] 杨波, 廖怡茗. 面向企业动态风险的知识图谱构建与应用研究[J]. 现代情报, 2021, 41(3): 110-120. [7] Bloem P, Wilcke X, van Berkel L, et al. kgbench: A collection of knowledge graph datasets for evaluating relational and m ultimodal machine learning [C].Proc of the 18th European Semantic Web Conf. Berlin: Springer, 2021: 614-630. [8] Zhang Y Y , Qian S S, Fang Q, et al. Multi-modal knowledge-aware attention network for question ans wering[J]. Journal of Computer Reserch and Development, 2020, 57(5): 1037-1045. [9] 江千军, 桂前进, 王磊, 等. 命名实体识别技术研究进展综述[J]. 电力信息与通信技术, 2022, 20(2): 15-24. [10] 曾青霞, 熊旺平, 杜建强, 等. 结合自注意力的BiLSTM-CRF的电子病历命名实体识别[J]. 计算机应用与软件, 2021, 38(3): 159-162. doi: 10.3969/j.issn.1000-386x.2021.03.024 [11] 陈茹, 卢先领. 融合空洞卷积神经网络与层次注意力机制的中文命名实体识别[J]. 中文信息学报, 2020, 34(8): 70-77. doi: 10.3969/j.issn.1003-0077.2020.08.009 [12] 史占堂, 马玉鹏, 赵凡, 等. 基于CNN-Head Transformer编码器的中文实体识别[J/OL]. 计算机工程: 1-10 [2022-03-16]. [13] Gligic L, Kormilitzin A, Goldberg P, et al. Named entity recognition in electronic health records using transfer learning bootstrapped neural networks[J]. Neural Networks, 2020, 121: 132-139. doi: 10.1016/j.neunet.2019.08.032 [14] 邱凌,张安思,李少波,等.航空制造知识图谱构建研究综述[J].计算机应用研究,2022,39(4):968-977. [15] 刘辉, 江千军, 桂前进, 等. 实体关系抽取技术研究进展综述[J]. 计算机应用研究, 2020, 37(S2): 1-5. [16] 陈晓玲, 唐丽玉, 胡颖, 等. 基于ALBERT模型的园林植物知识实体与关系抽取方法[J]. 地球信息科学学报, 2021, 23(7): 1208-1220. doi: 10.12082/dqxxkx.2021.200565