微博内容过滤技术的研究与实现

发布时间：2018-11-19 13:11

【摘要】：随着网络技术的蓬勃发展,越来越多的人开始使用微博。微博是一种信息传播媒体平台,在这个平台上,任何人都可以及时、快速、方便地发布自己的信息或者转发别人的信息,每天新产生的微博达数以万计传,不免其中某些微博包含一些不合法、不健康、不和谐的内容,所以需要一种系统能对微博的内容进行一定程度上的过滤处理,净化网络环境,确保给网民提供一个绿色、积极、健康的上网氛围。本文基于Java Web开发平台Myeclipse,使用SSH系统框架、MySQL数据库设计并实现微博内容过滤系统,首先本系统建立起一个微博平台,然后在此基础之上完成微博的过滤功能。本系统总体分为两块：前台系统和后台系统。前台系统主要是微博用户的功能,包括微博基本功能、微博内容过滤、用户信息设置、编辑关注者等功能模块；后台系统主要是系统管理员的功能,包括管理员管理、用户管理、资源管理、公告管理等功能。在这些功能模块中,重点设计实现的是微博内容过滤模块,该模块结合Lucene搜索引擎的查找方法与自然语言处理学科中的分词处理,完成微博内容过滤处理的两个功能：过滤屏蔽和过滤筛选。本系统开发以软件工程中的瀑布模型为设计主线,较为详细的介绍了本系统业务需求、功能需求、非功能需求、系统架构设计、功能设计和数据库设计。并对主要功能,给出了系统的实现环境、界面设计以及关键功能模块的代码实现过程,最后给出了系统的功能和性能测试结果。
[Abstract]:With the vigorous development of network technology, more and more people begin to use Weibo. Weibo is a media platform for the dissemination of information. On this platform, anyone can timely, quickly, and conveniently publish their own information or transmit other people's information. Inevitably, some of Weibo contains some illegal, unhealthy, and disharmonious content. Therefore, it is necessary to have a system that can filter the content of Weibo to a certain extent, purify the network environment, and ensure that Internet users are provided with a green, positive, A healthy online atmosphere. This paper uses SSH system framework and MySQL database to design and implement Weibo content filtering system based on Myeclipse, a Java Web development platform. Firstly, this system establishes a Weibo platform, and then accomplishes the filtering function of Weibo. This system is divided into two parts as a whole: foreground system and background system. The front desk system is mainly the function of Weibo user, including the basic function of Weibo, the content filtering of Weibo, the user information setting, editing the follower and so on. Background system is the main system administrator functions, including administrator management, user management, resource management, notice management and other functions. Among these functional modules, Weibo content filtering module is designed and implemented, which combines the search method of Lucene search engine with word segmentation in natural language processing. Complete Weibo content filtering two functions: filtering screening and filtering. The development of this system takes the waterfall model in software engineering as the main line of design. It introduces in detail the business requirements, functional requirements, non-functional requirements, system architecture design, functional design and database design of the system. For the main functions, the implementation environment, interface design and code implementation process of the key function modules are given. Finally, the function and performance test results of the system are given.
【学位授予单位】：厦门大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP311.52

【参考文献】