书信作为研究人物的重要资料，具有不可替代的史料价值和独特的文化价值。其作为编纂各种著作的原始材料，在特定的历史背景下可以提供一般官方文献中不具备的历史细节，因此近年来关于书信的梳理和释读研究逐渐增多。但问题也随之出现，在研究过程中，一封书信的研究往往需要大量的佐证材料以及与之关联的书信，但是由于知识组织的粗略化，对书信的整理缺乏细度和多层次的资源深加工，导致书信之间缺乏语义关联，造成了研究人员在查阅资料时需要多次查找，才能找到目标书信，耗时又耗力。因此如何实现书信及相关资源的快速搜集，是开展书信研究亟需解决的关键问题。

本文旨在解决目前书信研究过程存在的困难，深入挖掘书信的内容。通过研究实现书信资源的数字化存储与查询，同时揭示书信资源中更深层次的知识和关系、发现书信数据之间的内在逻辑关联、挖掘其中隐含的相关知识，因而提出构造书信本体，希望利用本体方法对书信进行规范化描述和深层次知识组织，实现书信的语义关联和检索，进而解决传统书信资源检索面临的问题。书目框架是最新的图书馆书目数据著录格式，其中的类和属性对于书信本体构建具有很大的借鉴意义。因此，本文选择以书目框架为基础，复用其中的一些类和属性，构造一个相对完善的书信本体。

本文从多方面对书信本体构建展开了研究。第一章主要是对于研究背景及国内外研究现状的阐述，目前国内外对于书信本体的专门研究较少，因此笔者调研了书目框架以及基于书目框架本体构建的研究现状，充分了解书目框架本体复用的情况，为实现书信本体构建奠定基础。第二章介绍了本体、书目框架的理论基础，重点介绍了本体构建的方法、构建工具以及书目框架的发展情况和优点。第三章通过分析书信关键要素，利用Protégé软件构建了书信本体，同时利用Jena推理机生成书信本体语义推理模型，之后定义了本体到数据库之间的映射规则，并将书信本体存储到图数据库Neo4j中。第四章开展了对于书信本体可用性的研究，设计了基于书信本体的语义检索系统并实现特定了功能，包括：主页面导航功能、关系检索及可视化、事件检索以及时空检索。实验结果显示，该系统基本实现了预定的目标与功能，验证了书信本体的可用性。

Letters have irreplaceable cultural value and historical value. They are not only important materials for studying a certain person, but also original materials for compiling various works. At the same time, they also have materials that are not available in general official documents, which can supplement some historical details. Therefore, in recent years, the research on combing and interpreting letters has increased gradually. However, problems also arise. In the process of research, a letter often needs a lot of supporting materials and letters related to it. However, due to the coarseness of knowledge organization, the lack of fineness and multi-level deep processing of resources, and the lack of semantic association between letters, it is time-consuming and labor-consuming to find the target by searching the data many times. How to realize the rapid collection of letters and related resources is the key problem of letter research.

In order to solve the problems existing in the study of letters and understand the contents of letters deeply, this paper not only realizes the digital preservation and search of letter resources, but also reveals the deeper knowledge and relationship in the letter resources, finds the internal logical relationship between the letter data and excavates the relevant knowledge hidden in it, so it proposes to construct the epistolary ontology, It is hoped that the ontology method can be used to standardize the description and deep-level knowledge organization of letters, to realize the semantic association and retrieval of letters, and then to solve the difficulties faced by traditional letter retrieval. The bibliographic framework is the latest library bibliographic data description format. The classes and attributes in the bibliographic framework are of great significance for the construction of the epistolary ontology. Therefore, this article chooses to use the bibliographic framework as the basis to reuse some of the classes and attributes to construct a relatively complete Epistolary ontology.

This paper studies the construction of epistolary ontology from many aspects. The first chapter is mainly about the research background and the research status at home and abroad. At present, there are few special researches on the epistolary ontology at home and abroad. Therefore, the research status of bibliographic framework and the construction of ontology based on bibliographic framework is investigated, and the reuse of bibliographic framework ontology is fully understood to lay a foundation for the construction of epistolary ontology. The second chapter introduces the theoretical basis of ontology and bibliographic framework, focusing on the methods and tools of ontology construction, as well as the development and advantages of bibliographic framework. In the third chapter, by analyzing the key elements of the letter, the epistolary ontology is constructed by using Protégé software, and the semantic reasoning model of the epistolary ontology is generated by Jena reasoning machine. Then, the mapping rules between the ontology and the database are defined, and the epistolary ontology is stored in the graph database Neo4j. In the fourth chapter, we study the usability of epistolary ontology, design a semantic retrieval system based on epistolary ontology and realize specific functions, including: main page navigation function, relationship retrieval and visualization, event retrieval and space-time retrieval. The experimental results show that the system basically achieves the predetermined goals and functions, and verifies the availability of the epistolary ontology.