【作者】 李建新；

【导师】 李卫华；

【作者基本信息】 广东工业大学 ， 计算机应用技术， 2011， 硕士

【摘要】 广东工业大学智能软件研究团队开发了一个以可拓策略生成方法为基础的自助游策略生成系统。该系统在解决经典旅游路线所需的时间与费用和人们实际能给予的时间与费用之间存在的矛盾上有着很好的优势。但是,通过对系统体系结构的分析,我们发现：原系统由于系统构件化程度低、体系结构不合理,线索过于单一化,导致了系统运行速度慢,而且对软件进行维护和功能扩充相当困难。本文针对以上所述的问题,提出适合于自助游策略生成系统的正交软件体系结构,利用正交软件体系结构线索性、层次性等特点,以可拓策略生成方法为基础,对系统进行重构,并在原系统基础上增加了两种旅游策略生成类型。由此,现有系统存在普通游、深度游、休闲游三种旅游策略生成类型。其中,普通游为旅游时间不超过一星期的旅游策略生成,该旅游策略生成线索的特点为旅游景点均为广州特色景点；深度游为旅游时间不超过两星期的旅游策略生成,相对于普通游类型,深度游策略生成增加了更多类型的旅游景点,扩展了用户的旅游范围；而当用户不满足于前两种旅游方式时,我们还提供了旅游时间不超过50天的休闲游策略生成。这样便满足了不同类型用户的需求。在系统存在问题的解决过程中,本文初步设计了一个适用于自助游策略生成系统的正交软件体系结构,提高了系统的性能,并为系统的后期维护与扩展提供了基础。本文首先介绍可拓策略生成、正交软件体系结构的国内外研究现状及本文研究的背景和意义,提出本文研究的主要内容。接着,对原自助游策略生成系统进行分析,总结出系统体系结构方面存在的主要问题。然后,提出了解决上述系统问题的自助游策略生成系统的正交化方案。最后,利用Eclipse开发平台和MySQL等开源工具,改进了自助游策略生成系统,并给出了系统的运行结果,以及新旧系统的性能对比结果。本文的创新之处：(1)在自助游策略生成系统特定的应用领域下,提出了适合于自助游策略生成系统的正交软件体系结构,提高了系统的可维护性和可扩展性；(2)对传统的正交软件体系结构进行了改进,增加了公共数据构件层,提高了系统性能,使其更加适合于自助游策略生成系统。本文是广东省自然科学基金资助项目(批准号：8151009001000041)—“软件Agent策略生成机制研究”和广东省自然科学基金资助项目(批准号10151009001000044)—“基于可拓数据挖掘的客户价值研究”的研究成果。更多还原

【Abstract】 The Intelligent Software Research Team of Guangdong University of Technology develops a self-help travel strategy generating system which takes Extenics Strategy Generating as basic theory. This system has a unique advantage in solving the conflict between the time-cost which the classic travel routes needed and the time-cost which people can give. But, after analyzing the software architecture of the self-help travel strategy generating system, we found that there are some problems. Such as components are not enough, the architecture is not reasonable and single clue in this system, which cause it running slowly and difficult to maintain and to expand functions.To solve these problems, this thesis proposes orthogonal software architecture which fits self-help travel strategy generating system, uses orthogonal software architecture which includes features such as thread and hierarchy and Extenics Strategy Generating to reconstruct this system. Based on the original system, we add two another travel strategy generating types. As a result, this system has ordinary travel, depth travel, and leisure travel totally three kinds of travel strategy generating types. Among them, the ordinary travel time for travel strategy does not generate more than one week, the travel strategy generating are mainly the characteristics of the travel attractions in Guangzhou; The depth travel time for travel strategy does not generate more than two week, relative to the ordinary travel type, depth travel strategy generating added more types of travel attractions, expands the scope of the user’s travel; And when the users still do not satisfy the previous two forms of travel, we also provide up to 50 days, leisure travel strategy generating. This will meet the needs of different types of users. In the process of system problems solution, this thesis preliminarily designs an orthogonal software architecture which fits the self-help travel strategy generating system. Then, we improve the performance of the system, and provide a basis to the later period development.Firstly, this thesis introduces the Extenics strategy generating, orthogonal software architecture research status and background and significance of this study, presents the main contents of this thesis. Secondly, this thesis analyzes the original strategy generating system, summarizes the main problems of the system architecture. Thirdly, the system is proposed to solve the above problem of orthogonal travel system program. Finally, this thesis uses Eclipse, MySQL and other open-source development platform and tools to improve the self-help travel strategy generating system. Then, this thesis gives the system’s operating results performance and comparison of the results of old and new systems.The innovation of this thesis:(1) In the specific areas of Self-help travel strategy generating system application, the thesis presents an orthogonal software system architecture which fit for the self-help travel generating system in order to improve the system’s maintainability and scalability;(2) The traditional orthogonal software architecture has been improved by increased the level of public data component, making it more suitable for self-help travel generating system.This is the research results of Guangdong Provincial Natural Science Foundation (Grant No.:8151009001000041)-"Research on the Strategy Generating Mechanism of Software Agents" and Guangdong Provincial Natural Science Foundation(Grant No.: 10151009001000044)-"The Client Values Research based on Extenics Data Mining".更多还原