面向我国畜牧业法律法规的知识图谱构建

doi:10.12160/j.issn.1672-5190.2023.03.010

摘要/Abstract

摘要：

[目的]提出一种对畜牧业法律法规知识组织的方法，使相关知识实现更加有序的组织。[方法]选取国家法律法规数据库、北大法宝、北大法意、威科先行和中国法律知识资源总库中的法律法规资源作为畜牧业法律法规的数据来源，对畜牧业相关法律法规进行筛选、获取，提取对应的制定机关、时效性、效力位阶、法规类别、公布日期、施行日期和年份信息，并对获得的数据进行预处理;对本体进行构建，定义实体和关系;基于规则的方法对获得的法律法规文本抽取实体和关系，并将抽取的实体和关系暂储到MySQL数据库中;利用Python中的SQLAlchemy库和py2neo库分别操作MySQL数据库和Neo4j图数据库，将MySQL中的数据转为实体关系三元组，存储到Neo4j图数据库中完成畜牧业法律法规知识图谱的存储。[结果]共收集到287个畜牧业相关的法律法规，按照法律法规是否含有章进行区分，其中，含有章的有211个，共1 470章，10 457条，不含章的76个，共2 145条;定义了“法律法规”“发布部门”“时效性”“效力级别”“法规类别”“章”“条”等共10种类型的实体，定义了“畜牧业法律法规-法律法规的类别关系”“畜牧业法律法规-法律法规的时效性关系”“畜牧业法律法规-法律法规的发布部门关系”“畜牧业法律法规-法律法规的效力级别关系”等共10种类型的关系;构建了一个含有实体数量14 936个、关系数量16 339个的畜牧业法律法规知识图谱。[结论]该方法可应用于畜牧业法律法规知识的组织，可使知识的组织更加有序，关联更加紧密。

关键词: 畜牧业, 知识图谱, 法律法规, Neo4j

Abstract:

[Objective] The present study was conducted to propose a method for organizing the knowledge of animal husbandry laws and regulations to achieve a more orderly organization of relevant knowledge. [Method] The online legal resources in the websites of flk.npc.gov.cn, pkulaw.com, lawyee.org, wkinfo.com.cn, and law.cnki.net were used as the data sources of animal husbandry laws and regulations, the relevant laws and regulations of animal husbandry were screened and obtained. Subsequently, the corresponding formulation authority, timeliness, effectiveness hierarchy, regulation category, issuing date, implementation date and year information were extracted, and the obtained data was preprocessed. The ontology was built, and the entities and relationships were defined. Using rule-based method, the entities and relationships from the obtained laws and regulations texts were extracted, and then temporarily stored in MySQL database. The SQLAlchemy database and py2neo database in Python were used to operate MySQL database and Neo4j Graph database to convert the data in MySQL into entity relationship triad, and then stored in Neo4j Graph database to complete the storage of knowledge graph of animal husbandry laws and regulations. [Result] A total of 287 laws and regulations related to animal husbandry were collected and classified based on whether the laws and regulations contained chapters. Among these, 211 laws and regulations contained chapters, totaling 1 470 chapters and 10 457 articles, whereas 76 laws and regulations did not contain chapters, totaling 2 145 articles. There were 10 types of entities defined, including ′laws and regulations′, ′issuing authority′, ′timeliness′, ′effectiveness hierarchy′, ′regulation category′, ′chapter′, ′article′, etc. There were 10 types of relationships defined, including ′the relationship of animal husbandry laws and regulations to category of laws and regulations ′, ′the relationship of animal husbandry laws and regulations to timeliness of laws and regulations′, ′the relationship of animal husbandry laws and regulations to issuing authority of laws and regulations′, and ′the relationship of animal husbandry laws and regulations to effectiveness hierarchy of laws and regulations′, etc. A knowledge graph of animal husbandry laws and regulations covering 14 936 entities and 16 339 relationships was constructed. [Conclusion] This method can be applied to organize the knowledge on animal husbandry laws and regulations, making the organization of knowledge more systematic and closely related.

Key words: animal husbandry, knowledge graph, law and regulation, Neo4j

中图分类号:

张容祯, 孟小艳, 刘潇潇, 汪洋. 面向我国畜牧业法律法规的知识图谱构建[J]. 畜牧与饲料科学, 2023, 44(3): 69-74.

ZHANG Rongzhen, MENG Xiaoyan, LIU Xiaoxiao, WANG Yang. Construction of Knowledge Graph for Animal Husbandry Laws and Regulations in China[J]. Animal Husbandry and Feed Science, 2023, 44(3): 69-74.

图/表 6

图1

表1

表2

表3

表4

表5

参考文献 20

[1]	数字赋能助推智慧畜牧业发展[J]. 中国畜牧业, 2022(15):8.
[2]	戴高阳. 面向畜牧业的知识图谱构建技术研究[D]. 乌鲁木齐: 新疆农业大学, 2022.
[3]	杜何哲. 融合知识图谱的林业法律法规问答系统研建[D]. 北京: 北京林业大学, 2020.
[4]	PUJARA J, MIAO H, GETOOR L, et al. Knowledge graph identification[C]// International Semantic Web Conference. Berlin, Heidelberg: Springer, 2013:542-557.
[5]	刘峤, 李杨, 段宏, 等. 知识图谱构建技术综述[J]. 计算机研究与发展, 2016, 53(3):582-600.
[6]	徐增林, 盛泳潘, 贺丽荣, 等. 知识图谱技术综述[J]. 电子科技大学学报, 2016, 45(4):589-606.
[7]	SUCHANEK F M, KASNECI G, WEIKUM G. YAGO: A large ontology from wikipedia and WordNet[J]. Journal of Web Semantics, 2008, 6(3):203-217. doi: 10.1016/j.websem.2008.06.001
[8]	BIZER C, LEHMANN J, KOBILAROV G, et al. DBpedia - A crystallization point for the Web of Data[J]. Journal of Web Semantics, 2009, 7(3):154-165. doi: 10.1016/j.websem.2009.07.002
[9]	JI L, WANG Y J, SHI B T, et al. Microsoft concept graph: Mining semantic concepts for short text understanding[J]. Data Intelligence, 2019, 1(3):238-270. doi: 10.1162/dint_a_00013
[10]	FELLBAUM C. WordNet[M]//POL R, HEALY M, KAMEAS A. Theory and applications of ontology: Computer applications. Dordrecht:Springer, 2010:231-243.
[11]	田玲, 张谨川, 张晋豪, 等. 知识图谱综述:表示、构建、推理与知识超图理论[J]. 计算机应用, 2021, 41(8):2161-2186. doi: 10.11772/j.issn.1001-9081.2021040662
[12]	周辉. 基于农业知识图谱的问答系统设计与研究[D]. 常州: 常州大学, 2022.
[13]	许思特, 孙木. 基于命名实体识别与Neo4j的中文电子病历知识图谱构建和应用[J]. 医学信息学杂志, 2022, 43(12):50-56.
[14]	王天一, 孟小亮, 张华. 一种面向生态环境领域的知识图谱构建方法[J]. 地理空间信息, 2023, 21(1):14-19.
[15]	胡杰, 李源洁, 耿號, 等. 基于深度学习的汽车故障知识图谱构建[J]. 汽车工程, 2023, 45(1):52-60,85.
[16]	国家图书馆研究院. 国家法律法规数据库正式上线[J]. 国家图书馆学刊, 2021(2):79.
[17]	李韵石. 北大法宝:探寻“智慧法治”[J]. 法人, 2021(11):83-85.
[18]	程雪艳. 国内两大中文法律数据库比较研究[J]. 河南图书馆学刊, 2007, 27(4):10-12.
[19]	陈爽. 荷兰威科集团发布“威科先行法律信息库”[J]. 计算机与网络, 2015, 41(10):79.
[20]	中国法律知识资源总库:规范、完备、智能、整合的法律学习应用平台[J]. 法制资讯, 2014(8):97.