[目的/意义] 知识图谱已成为公共数字文化资源知识组织的新形态。利用知识图谱技术赋能红色历史人物知识问答服务，提升用户交互体验，对红色历史资源的开发利用具有重要意义。[方法/过程] 在梳理历史人物数字资源组织及知识问答系统相关研究的基础之上，构建了红色历史人物知识图谱Schema与KBQA架构，从数据获取、知识抽取、知识融合、图谱生成和知识问答五个环节搭建了红色历史人物问答模型，并以老上大历史人物数字资源进行实证研究。[结果/结论] 本文设计的知识问答服务架构在红色历史人物数字资源的图谱半自动构建、知识推理与智能交互方面具有优越性，提升了用户知识服务体验。
[Purpose/significance] Knowledge graph has become a new form of public digital cultural resources organization. Using knowledge graph technology to enable the Knowledge Q & A service of red historical figures and improve user interaction experience is of great significance to the development and utilization of red historical resources.[Method/process] On the basis of combing the related research of digital resource organization and Knowledge Q & A system of historical figures, the paper constructed the knowledge graph schema and KBQA architecture of red historical figures, and then built the model of Q & A from five aspects of data acquisition, knowledge extraction, knowledge fusion, graph generation and Knowledge Q & A.This paper took the red historical figures digitalresources of Shanghai University(1922-1927) as an example for empirical research.[Result/conclusion] The Knowledge Q & A service architecture designed in this paper has advantages in semi-automatic graph construction, knowledge reasoning and intelligent interaction of digital resources of red historical figures, and improves the user knowledge service experience.
[1] 中共中央办公厅,国务院办公厅.关于实施革命文物保护利用工程(2018-2022年)的意见[N].人民日报,2018-07-30(001).
[2] Harvard University, Academia Sinica, Peking University. China biographical database[EB/OL].[2021-03-03]. https://projects.iq.harvard.edu/cbdb.
[3] 张文勇.湖南近代人物数据库建设研究与实现[D].长沙:中南大学,2013.
[4] 靳志军,何寿峰,郝春柳,等.河北红色历史文化资源挖掘研究-以李大钊专题数据库建设为例[J].文化月刊,2016(7):114-115.
[5] 刘宁静,刘音,王莫言,等.数字人文视角下学术名人知识模型构建研究——以李政道数字资源中心为例[J].图书情报工作,2019,63(23):113-121.
[6] 陈涛,刘炜,单蓉蓉,等.知识图谱在数字人文中的应用研究[J].中国图书馆学报,2019,45(6):34-49.
[7] 潘俊.面向数字人文的人物分布式语义表示研究——基于CBDB数据库和古籍文献[J].图书馆杂志,2020,39(8):94-102.
[8] 杨海慈,王军.宋代学术师承知识图谱的构建与可视化[J].数据分析与知识发现,2019,3(6):109-116.
[9] 周亦,周明全,王学松,等.大数据环境下历史人物知识图谱构建与实现[J].系统仿真学报,2016,28(10):2560-2566.
[10] 单良,刘欣.基于中国历史人物知识的智能问答系统构建[J].情报探索,2019(6):101-105.
[11] COCCO R, ATZORI M, ZANIOLO C, etal. Machine learning of SPARQL templates for question answering over LinkedSpending[J].CEUR workshop proceedings,2019,2400:156-161.
[12] 李贺,刘嘉宇,李世钰,等.基于疾病知识图谱的自动问答系统优化研究[J].数据分析与知识发现,2021,5(5):115-126.
[13] 陈璟浩,曾桢,李纲.基于知识图谱的"一带一路"投资问答系统构建[J].图书情报工作,2020,64(12):95-105.
[14] 高劲松,方晓印,刘思洋,等.基于关联数据的馆藏文物资源知识关联与智能问答研究[J].情报科学,2021,39(5):12-20.
[15] WU W Q,ZHU Z F,LU Q, etal. Introducing external knowledge to answer questions with implicit temporal constraints over knowledge base[J].Future Internet,2020,12(3):45.
[16] 谢项.基于古诗知识图谱的智能问答研究[D].武汉:华中师范大学,2020.
[17] YU J,ZHU Z H,WANG Y J, etal. Cross-modal knowledge reasoning for knowledge-based visual question answering[EB/OL].[2021-08-02].https://www.sciencedirect.com/science/article/pii/S0031320320303666?via%3Dihub.
[18] HUANG T S,LI X W,ZHAI S P,etal. Knowledge graph reasoning based on tensor decomposition and MHRP-Learning[EB/OL].[2021-08-02].https://downloads.hindawi.com/journals/am/2021/8880553.pdf.
[19] TANG H,DONG H J,ZHOU Q J. End-to-end masked graph-based CRF for joint slot filling and intent detection[J].Neurocomputing,2020,413(6):348-35.
