基于本体的版权知识库构建方法研究与应用

发布时间：2018-05-13 07:51

本文选题：版权 + 本体　；参考：《北方工业大学》2017年硕士论文

【摘要】：本体是源自哲学的一个概念,近年来随着计算机信息科学的发展被引入到计算机领域,用于构建所描述事物的规范概念或术语,使得机器可以理解知识。本体在人工智能和语义网的发展中有重要的作用。对本体的研究已成为知识库构建方法研究的重要组成部分。版权[49](又称著作权),是作者对作品依法享有的财产权和人身权等。版权是作者对他的作品依法享有的复制权利的合法所有权,包括计算机程序、文学作品、音乐视频作品(如音乐、电影)以及摄影摄像作品。基于本体的版权知识库是通过本体技术将版权知识进行组织、整理以及建立概念之间的关系网格,并提供检索服务的知识集群。传统的知识库检索手段通过简单的字符串匹配方式进行检索,导致知识获取效率低下。近年来采用语义检索方式己成为提高检索效率的重要方法。本文根据本体的特性,设计出基于本体的版权知识库构建模型,使得版权知识库具备语义性质,可以提供具有语义推理功能的检索服务。在本体构建方法上,本文针对版权案例文本属于半结构化知识,可以用自动构建的方式构建本体,而文本量小且准确性要求较高的法律知识需要手动构建本体。该方法可以实现自动构建本体,减少构建时间。传统知识库不能满足人们日益增长的知识需求,因此课题提出利用本体的语义特性的改进传统知识库,使知识库可以理解知识语义。本文针对版权知识的结构特点对本体构建方法进行了改进,提出了基于本体的版权知识库构建模型,该模型包括三个方面:(1)知识库的规划与设计。(2)本体构建。(3)知识服务(可视化展示和检索服务)。本课题使用图形化的本体构建工具Protege来构建本体框架,然后利用Jena本体操作工具实现本体实例化和知识推理。知识检索功能通过开源全文检索引擎工具包Lucene实现的,通过Jena本体技术与Lucene全文检索技术的融合提高查询准确率。最后,通过构建基于本体的版权知识检索服务系统,验证基于本体的版权知识库构建方法的可行性。
[Abstract]:Ontology is a concept derived from philosophy. In recent years, with the development of computer information science, it has been introduced into the field of computer, which is used to construct normative concepts or terms for describing things, so that machines can understand knowledge. Ontology plays an important role in the development of artificial intelligence and semantic web. The research of ontology has become an important part of knowledge base construction. Copyright [49] (also known as copyright) is the author's property right and personal right in accordance with the law. Copyright is the legal right of the author to copy his works, including computer programs, literary works, music and video works (such as music, movies) and photographic and video works. The copyright knowledge base based on ontology is a knowledge cluster that organizes, arranges and builds the relational grid of concepts and provides retrieval services through ontology technology. The traditional method of knowledge base retrieval is simple string matching, which leads to low efficiency of knowledge acquisition. In recent years, semantic retrieval has become an important method to improve retrieval efficiency. According to the characteristics of ontology, this paper designs a model of copyright knowledge base based on ontology, which makes copyright knowledge base have semantic properties and can provide retrieval services with semantic reasoning function. In terms of ontology construction method, the text of copyright case belongs to semi-structured knowledge, which can be constructed automatically. However, the legal knowledge with small amount of text and high accuracy needs to build ontology manually. This method can automatically build ontology and reduce the construction time. The traditional knowledge base can not meet the increasing demand of knowledge, so we propose to improve the traditional knowledge base by using the semantic characteristics of ontology, so that the knowledge base can understand the semantics of knowledge. In this paper, according to the structural characteristics of copyright knowledge, the method of ontology construction is improved, and the model of building copyright knowledge base based on ontology is proposed. The model includes three aspects: (1) Planning and design of knowledge base. (2) Ontology construction. (3) knowledge service (visual presentation and retrieval service). In this paper, we use the graphical ontology construction tool Protege to construct the ontology framework, and then use the Jena ontology operation tool to realize ontology instantiation and knowledge reasoning. The knowledge retrieval function is realized by open source full-text search engine (Lucene), and the query accuracy is improved by the fusion of Jena ontology technology and Lucene full-text retrieval technology. Finally, the ontology-based copyright knowledge retrieval service system is constructed to verify the feasibility of the ontology-based copyright knowledge base construction method.
【学位授予单位】：北方工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.3

【参考文献】