联系我们

论文查询结果	返回搜索

论文编号：	14638
作者编号：	2120223716
上传时间：	2024/6/5 16:12:37
中文题目：	少数民族节日文化知识图谱构建研究 ——以贵州苗族节日文化为例
英文题目：	A Study on the Construction of a Knowledge Map of Ethnic Minority Festival Culture: Taking Guizhou Miao Festival Culture as an Example
指导老师：	冯湘君
中文关键字：	贵州苗族节日文化；领域本体；通用本体；知识图谱；信息组织
英文关键字：	Guizhou Miao Festival Culture; Domain Ontology;Universal Ontology; Knowledge Graph; Information Organization
中文摘要：	贵州苗族的节日是我国传统文化的核心构成之一，历史底蕴深厚、民俗艺术丰富，作为我国多元文化的重要组成部分，其文化遗产的价值不容忽视。然而，随着现代化的快速发展和外来文化的冲击，少数民族文化的保护与传承面临巨大挑战。由于少数民族文化数据具有异构性、分散性和复杂性等特点，其保护工作尤为艰巨。知识图谱作为人工智能领域的关键技术，在构建结构化知识库、组织管理知识以及识别知识等方面具有显著优势。因此，利用知识图谱技术对贵州苗族文化进行数字化保护与传承具有广阔的应用前景。鉴于此，本文的核心研究内容共分为三个部分。 （一）贵州苗族节日文化本体构建研究。针对贵州苗族节日文化领域本体构建方法的欠缺，本文从贵州苗族节日文化实际应用场景出发，深入挖掘贵州苗族节日文化的核心特点。通过综合分析田野调查所获相关数据资料以及各类文献资料，结合七步法等现有本体构建方法的优点，提出一套适用于贵州苗族节日文化领域的本体构建方案。依据该方案，本文利用本体描述语言精确描述贵州苗族节日文化的内在逻辑和层次结构。并利用 Protégé工具进行本体存储，为后续构建贵州苗族节日文化知识图谱奠定坚实基础。 （二）贵州苗族节日文化知识图谱构建。针对贵州苗族节日文化内的非结构化数据的知识抽取难题，本文提出了注意力机制与 BERT-BiLSTM-CRF 模型结合的实体抽取方法，注意力机制的应用使得该模型对贵州苗族节日文化字词上下文关注得到提高，从而尽可能规避了无关信息对实体识别的影响，最后采用 BERT-BiLSTM-Att-CRF 模型对实体中关系进行提取，并将成功提取的贵州苗族节日文化实体关系以三元组形式表示，用 Neo4j 图数据库进行存储，以此完成贵州苗族节日文化知识图谱的构建。 （三）贵州苗族节日文化知识图谱应用展示。在前文贵州苗族节日文化知识图谱的基础上，进一步利用 Flask 框架等前端可视化技术，将该知识图谱转化为一个交互式、直观的应用服务。通过直观的图形展示和丰富的交互功能，并对其进行严格的测试。测试结果表明，该应用能够根据用户输入的检索数据，准确展示相关信息，为用户提供准确、全面的贵州苗族节日文化知识。
英文摘要：	The festival of the Miao nationality in Guizhou is the core component of China's traditional culture, containing profound historical heritage and rich folk art, as an important cornerstone of China's multiculturalism, the value of its cultural heritage cannot be ignored. However, with the rapid development of modernization and the impact of foreign cultures, the protection and inheritance of ethnic minority cultures are facing great challenges. Due to the heterogeneity, dispersion and complexity of ethnic minority cultural data, its protection is particularly arduous. As a key technology in the field of artificial intelligence, knowledge graph has significant advantages in building a structured knowledge base, organizing and managing knowledge, and identifying knowledge. Therefore, the use of knowledge graph technology to digitally protect and inherit the Miao culture in Guizhou has broad application prospects. In view of this, the core research content of this paper is divided into three parts. (1) Research on the construction of the ontology of Miao festival culture in Guizhou. In view of the lack of ontology construction methods in the field of Miao festival culture in Guizhou, this paper starts from the practical application scenarios of Guizhou Miao festival culture and deeply excavates the core characteristics of Guizhou Miao festival culture. Based on a comprehensive analysis of the relevant data obtained from field surveys and field interviews, as well as various literatures, combined with the advantages and disadvantages of existing ontology construction methods such as the seven-step method, a set of ontology construction schemes suitable for the field of Miao festival culture in Guizhou were proposed. According to this scheme, this paper uses ontology description language to accurately describe the internal logic and hierarchy of Miao festival culture in Guizhou. The Protégé tool was used to store the ontology, which laid a solid foundation for the subsequent construction of Guizhou Miao festival culture knowledge map. (2) Construction of Guizhou Miao festival cultural knowledge map. Most of thedata resources in Guizhou Miao festival culture are unstructured, so the knowledge extraction of unstructured Guizhou Miao festival culture data is the core content and difficulty of this paper. In order to solve this problem, this paper proposes an entity extraction method combining the attention mechanism and the BERT-BiLSTM-CRF model, and the application of the attention mechanism improves the model's attention to the context of Guizhou Miao festival cultural words, so as to avoid the influence of irrelevant information on entity recognition as much as possible. Finally, the BERT- BiLSTM-Att-CRF model is used to extract relationships within entities, and represent the successfully extracted entity relationships of Guizhou Miao festival culture in triplet form, store them in Neo4j graph database, and thus complete the construction of Guizhou Miao festival culture knowledge graph. (3) Guizhou Miao festival cultural knowledge map application display. On the basis of the knowledge graph of Guizhou Miao festival culture mentioned above, the front-end visualization technology such as Flask framework is further used to transform the knowledge graph into an interactive and intuitive application service. Through intuitive graphical presentation and rich interactive functions, it is rigorously tested. The test results show that the application can accurately display relevant information according to the retrieval data entered by users, and provide users with accurate and comprehensive cultural knowledge of Guizhou Miao festivals
查看全文：	预览  下载(下载需要进行登录)