论坛用户行迹分析系统的设计与实现

发布时间：2018-05-19 04:09

本文选题：网络论坛 + 用户行迹　；参考：《哈尔滨工业大学》2017年硕士论文

【摘要】：随着互联网与人类生活进一步融合,出现了各种各样的网络应用,如在线论坛、电子商务、社交软件、网络游戏等。互联网在为人类生活提供便捷的同时也由于其虚拟性带来了诸多问题。近年来互联网向金融领域的扩展加速了网络实名制的进程,推进了可信网络空间建设。但网络论坛由于其讨论交流的定位及非企业法人维护等原因,用户在网络论坛中依然使用着虚拟身份,为许多网络违法行为提供了藏匿空间。对网络中虚拟身份背后社会主体的追溯成为一个被关注的问题。针对这一问题,本文基于网络用户命名习惯性与中小型论坛用户同一性,提出一种通过发现用户在互联网论坛空间内活动行迹进而挖掘虚拟身份背后社会主体信息的方法。其中网络用户命名习惯性指网络用户在互联网使用中在多个网络应用或站点中使用相同的id进行账号注册。中小型论坛用户的同一性是指这些网络论坛中聚集的用户具有相同的特征。本文中所讨论的用户虚拟身份的标识包括邮箱和用户名。首先,本文通过链接扩展和站点类型识别来发现当前互联网中的中文论坛站点从而构建用户论坛活动行迹的“地图”。其次,通过对站点注册查重接口的模拟来发现用户的活动论坛集合,再基于论坛内容爬虫记录获取用户在每个论坛内的发回帖信息,发现用户的论坛活动行迹。随后对所发现的用户行迹进行分析,从用户的发回帖记录中匹配邮箱、手机号等个人信息,基于站点类别粗粒度的定位用户关注领域,并根据同领域站点注册数量、站点规模、单一站点用户活跃度、单一站点用户影响力等信息来量化用户的领域影响力和领域兴趣度。最后,本文将用户行迹分析作为一种服务基于Web Service构建行迹分析平台来提供服务获取接口。并在此基础上对Web接口的输入输出进行可视化封装来实现系统前端。
[Abstract]:With the further integration of the Internet and human life, a variety of network applications have emerged, such as online forums, electronic commerce, social software, online games and so on. Internet not only provides convenience for human life, but also brings many problems because of its virtual nature. In recent years, the expansion of the Internet to the financial field has accelerated the process of network real name system and promoted the construction of trusted cyberspace. However, due to the location of discussion and communication and the maintenance of non-corporate legal person, users still use virtual identity in the network forum, which provides hiding space for many illegal activities on the network. The tracing of the social subject behind the virtual identity in the network has become a concerned issue. In order to solve this problem, based on the identity of network users' naming habits and small and medium-sized forum users, this paper proposes a method to mine the information of social subjects behind virtual identity by discovering users' activities in the Internet forum space. Network user naming habit means that network users use the same id to register their accounts in multiple network applications or sites in the use of the Internet. The identity of the users of small and medium-sized forums means that the users gathered in these web forums have the same characteristics. The identity of the user's virtual identity discussed in this article includes a mailbox and a user name. First of all, this paper uses link extension and site type identification to find out the Chinese forum sites in the current Internet to construct a "map" of user forum activity. Secondly, the user's active forum set is found through the simulation of the site registration and duplicate interface, and then the user's post information in each forum is obtained based on the forum content crawler record, and the user's forum activity track is found. Then we analyze the user's tracks, match the personal information such as mailbox, mobile phone number and other personal information from the user's post record, locate the user's domain of concern based on the coarse-grained site category, and according to the number of sites registered in the same domain, site size, etc. Single site user activity, single site user influence and other information to quantify users' domain influence and domain interest. Finally, this paper uses user trace analysis as a service based on Web Service to build a platform to provide service acquisition interface. On the basis of this, the input and output of Web interface are encapsulated visually to realize the front end of the system.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP393.09

【参考文献】