RSS资源聚合系统搜索引擎的设计与实现

发布时间：2018-03-01 18:45

本文关键词： RSS 资源聚合评价算法搜索模型搜索引擎　出处：《中山大学》2012年硕士论文　论文类型：学位论文

【摘要】：互联网信息的爆发式增长给人们带来了信息量过载的问题，用户在庞大的网络资源中获取自己关注的信息已经越来越困难。RSS信息聚合平台为用户提供了多样化、个性化的信息获取方式，便于用户及时快速地获取自己关注的网络信息。随着RSS信息聚合平台中数据源的不断更新，平台中的数据量也日渐庞大。当数据量远远超过用户逐一浏览信息的能力范围时，平台中的搜索引擎就成为了用户精确检索信息的主要工具。本文围绕着实现一个RSS资源聚合系统的搜索引擎，通过研究系统中RSS数据源的特点，提出了面向RSS资源聚合系统的搜索模型。该模型在具备了现有RSS信息聚合平台基本搜索功能的基础上，结合了Lucene排序算法和RSS数据源评价算法，使得其搜索的排序结果更加贴近于系统中RSS数据源的特点，，为RSS资源聚合系统提供更贴切其特征的搜索服务。本文完成的主要工作如下：首先，本文提出了RSS数据源评价算法。本文分析了RSS资源聚合系统中数据源的特点，借鉴网页链接分析中的排序算法思想，提出了RSS数据源的评价算法。通过该算法可以衡量系统中不同RSS数据源之间质量的高低。其次，本文结合了Lucene排序算法和RSS数据源的评价算法，设计了一个面向RSS资源聚合系统的搜索模型。该搜索模型加强了RSS数据源质量和条目发布时间对搜索排序结果的影响，使得排序结果更加符合系统中数据的特征，为系统提供更贴切的搜索服务。本文通过实验和测试验证了该模型的有效性和可行性。最后，本文使用检索引擎工具包Lucene构建一个搜索引擎的框架，实现了“HCI@RSS”资源聚合系统的搜索引擎。该搜索引擎应用了面向RSS资源聚合系统的搜索模型。本文最后遵循软件工程的方法描述了系统搜索引擎模块的设计和实现过程，并通过系统的测试和运行验证了该搜索引擎的有效性和可行性。
[Abstract]:The explosive growth of Internet information brings people the problem of information overload. It is more and more difficult for users to obtain the information of their concern in the huge network resources. RSS information aggregation platform provides users with a variety of information. Personalized information acquisition method is convenient for users to get their concerned network information in time and quickly. With the continuous updating of the data sources in the RSS information aggregation platform, the amount of data in the platform is becoming larger and larger. When the amount of data is far beyond the range of users' ability to browse the information one by one, The search engine in the platform has become the main tool for users to retrieve information accurately. This paper focuses on the implementation of a RSS resource aggregation system search engine, through the study of the characteristics of RSS data sources in the system, This paper presents a search model for RSS resource aggregation system, which combines the Lucene sorting algorithm and the RSS data source evaluation algorithm on the basis of the basic searching function of the existing RSS information aggregation platform. It makes the search result more close to the characteristics of the RSS data source in the system, and provides a more appropriate search service for the RSS resource aggregation system. The main work of this paper is as follows:. First of all, this paper puts forward the RSS data source evaluation algorithm, analyzes the characteristics of the data source in the RSS resource aggregation system, and draws lessons from the idea of sorting algorithm in the web link analysis. The evaluation algorithm of RSS data sources is proposed, by which the quality of different RSS data sources in the system can be measured. Secondly, combining the Lucene sorting algorithm and the RSS data source evaluation algorithm, this paper designs a search model for RSS resource aggregation system, which strengthens the influence of RSS data source quality and item publishing time on the search sorting results. The sorting results are more consistent with the characteristics of the data in the system and provide a more appropriate search service for the system. The validity and feasibility of the model are verified by experiments and tests. Finally, this article uses Lucene, a search engine toolkit, to build a framework for a search engine. The search engine of "HCI@RSS" resource aggregation system is implemented. This search engine applies the search model oriented to RSS resource aggregation system. Finally, the design and implementation process of search engine module is described according to the method of software engineering. The effectiveness and feasibility of the search engine are verified by testing and running of the system.
【学位授予单位】：中山大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP391.3

【相似文献】