面向数字记忆开发利用的档案检索模型构建研究  

Research on the Construction of Archives Retrieval Model for the Digital Memory Development and Utilization

作　　者：房小可[1] FANG XiaoKe(School of Applied Arts and Science,Beijing Union University,Beijing 100191,P.R.China)

机构地区：[1]北京联合大学应用文理学院,北京100191

出　　处：《数字图书馆论坛》2021年第11期21-27,共7页Digital Library Forum

基　　金：国家社会科学基金青年项目“面向社会记忆构建的档案资源检索研究”(编号:18CTQ041)资助。

摘　　要：本文从构建目标、数字记忆开发利用途径和档案组织粒度三方面探讨数字记忆和档案检索之间的逻辑关系,并在此基础上构建面向数字记忆开发利用的档案检索模型。模型主要分为档案信息中数字记忆要素提取、要素语义关系提取和索引库建立及匹配。其中数字记忆要素提取分为基于命名实体识别方法的实体要素提取,以及基于LDA模型的主题提取;要素语义关系提取分为基于神经网络的实体关系提取和基于空间向量相似性的主题关系提取;索引库建立及匹配模块旨在通过检索数字记忆要素字段获取档案承载记忆的基因链,实现记忆的完整再现,促进档案价值的开发利用。This paper discusses the logical relationship between digital memory and archives retrieval model from the aspects of the goal of construction,the way of digital memory utilization and the granularity of archives organization.Then,this pater constructs archives retrieval model oriented to digital memory utilization.The model is mainly divided into three parts:the extraction of digital memory elements,the extraction of semantic relationship of elements,and the establishment of index database and resource matching.Digital memory feature extraction includes entity feature extraction based on named entity recognition method and topic extraction based on LDA model;Feature semantic relation extraction includes entity relation extraction based on neural network and topic relation extraction based on spatial vector similarity;The index database building and matching module aims to retrieve the digital memory element fields to obtain the gene chain of memory carried by archives,which realizes the complete reproduction of memory,and promotes the development and utilization of value of archives.

关 键 词：数字记忆 信息检索模型 数字记忆要素 语义关联 

分 类 号：G270.7[文化科学—档案学]