【作者】 贺鑫；

【导师】 林正奎；

【作者基本信息】 大连海事大学 ， 计算机科学与技术， 2010， 硕士

【摘要】 随着信息技术的迅猛发展,数据库技术已经渗透到经济和社会的各个领域,并已被公认为是信息资源开发、管理和服务最有效的手段之一。其中数据库设计作为建立数据库及其应用系统的基础和核心技术尤其得到了社会各界的广泛关注,并已成为数据库技术领域研究的重点内容。按照规范化设计方法,数据库设计通常划分为需求分析、概念结构设计、逻辑结构设计、物理结构设计、数据库实施、运行和维护六个阶段。其中,概念结构设计是通过对用户需求进行综合、归纳与抽象,形成一个独立于具体数据库系统的概念数据模型的过程。作为联系现实世界与机器世界的桥梁和纽带,概念结构设计,尤其是概念数据建模的质量与效能直接决定了数据库系统建设的成败,因此在数据库设计中一直占有非常重要的地位。本文针对目前概念数据建模主要采用手工设计方法,建模质量和效能过分依赖设计人员个人经验和水平的实际情况,提出了一种基于用户报表的概念数据模型快速建模方法,旨在通过对各类复杂用户报表共性结构和语义特征的分析,实现用户报表所蕴含实体及其数据结构信息的自动提取,并将其自动转化为E-R模型,以提高概念数据模型的建模效率。主要研究工作包括：复杂用户报表的规范化处理与统一形式化描述方法、用户报表与E-R模型的概念映射体系、用户报表数据项提取及分类算法、用户报表数据结构特征提取算法,以及用户报表数据结构特征到E-R模型的转换方法。在此基础之上,通过开发一个原型系统,对研究成果的正确性和有效性进行了初步验证。概念数据模型自动化建模技术是一项极具挑战性的研究工作,由于时间及自身水平有限,论文研究工作尚缺乏深度,且存在很多不足之处,作者将在后续的研究过程中加以持续改进。更多还原

【Abstract】 With the rapid development of information technology, database technology has permeated every realm of economy and society, and has been generally acknowledged as one of the most effective methods of the information resource development, management and service. As the foundation and core technology of building database and data base application system, database design has particularly drawn wide attention from various circles of society, and has become key research content in database technology field.According to the standardized design, database design is usually divided into six stages, including requirement analysis, conceptual structure design, logical structure design, physical structure design, database implementation, operation and maintenance. Among them, conceptual structure design is the process forming a conceptual data model independent of a specific database system through synthesis, induction and abstraction of user requirements. As a bridge and link of the real world and the machine world, conceptual structure design, particularly the quality and efficacy of the conceptual data modeling determines the success or failure of the construction of database system. Therefore, conceptual structure design plays a very important role in the database design.Aiming at the present situation that conceptual data modeling mainly by manual designing, and which quality and effectiveness over-reliance on designers’ personal experience and standard, this paper proposes a rapid modeling method of conceptual data model based on user views. It can improve the efficiency of conceptual data modeling by analysis various complex common structural and semantic features, to realize the automatic extraction of entities implied in user views and their data structure information, and then convert to ER model automatically. Main research activities include:the normalized treatment of complex user views and unified formal description method, the concept mapping system between user views and E-R model, extraction and classification algorithm for data items of user views, and the conversion method from data structure features of user views to the ER model. On this basis, It preliminary verifies the accuracy and validity of research conducted by developing a prototype system. Conceptual data model automated modeling technology is a challenging research work, due to time and own standard limited, this paper studies still lack depth and exist a great deal of insufficient, there will be continuous improvement in the follow-up research process.更多还原