基于BETR-BiGRU-CRF模型的文物档案知识图谱构建研究——以北京中轴线文化遗产档案为例
Research on Knowledge Graph Construction of Cultural Heritage Archives Based on BETR-BiGRU-CRF Model——Through Taking Beijing Central Axis Cultural Heritage Archives as an Example
关定邦 1赵庆聪2
作者信息
- 1. 北京信息科技大学信息管理学院,北京 100192
- 2. 北京信息科技大学信息管理学院,北京 100192;绿色发展大数据决策北京市重点实验室,北京 100192
- 折叠
摘要
建立文物档案知识图谱便于对文物档案进一步的研究与利用.以中轴线文化遗产档案为原始数据集,采用BERT-BiGRU-CRF模型进行文物档案实体识别,在此基础上构建中轴线文化遗产档案知识图谱,使用Neo4j图数据库完成知识存储.经实验验证,BERT-BiGRU-CRF模型在文物档案实体识别任务中具有更高的精确度与召回率,有助于建立准确、一致、完整的文物档案知识图谱,能够为文物档案管理与保护、文化遗产保护与传承等提供参考与借鉴.
Abstract
Establishing a knowledge map of cultural heritage archives is beneficial to further research and utilization of cultural heritage archives.The BERT-BiGRU-CRF model is used to identify the heritage archives entities with the original dataset of the cultural heritage archives of the central axis,on which the knowledge graph of the cultural heritage archives of the central axis is constructed,and the knowledge storage is completed with Neo4j graph database.It is experimentally verified that the BERT-BiGRU-CRF model has higher accuracy and recall in the task of heritage archive entity recognition,which helps to build an accurate,consistent and complete knowledge map of heritage archives.This can provide reference for cultural relics archives management and protection,cultural heritage protection and inheritance,etc.
关键词
文物档案/知识图谱/实体抽取/BETR-BiGRU-CRFKey words
Heritage archives/Knowledge graph/Entity extraction/BETR-BiGRU-CRF引用本文复制引用
基金项目
中轴线保护中心改造建设项目中轴线档案建设项目()
出版年
2024
NETL
NSTL
万方数据