Dissertation
Dissertation > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Computer network > General issues > Computer Network Security

Recognition and Judgment of Public Relations Online on Bulletin Board System

Author GaoZuo
Tutor KangYanMei
School Institute of International Relations
Course Communication and Information System
Keywords Public Relations Online Recognition Content Security Classification Feature Selection
CLC TP393.08
Type Master's thesis
Year 2012
Downloads 4
Quotes 0
Download Dissertation

The existence of public relations online brings about an increasing number ofcontent security problems. And inappropriate contents from public relations online,such as slanders, false information and illegal contents adversely affect the networkenvironment and damage social and economic orders. Recognition and judgment ofpublic relations online has become an urgent problem in the context of networkpopularization.To solve the content security problems caused by public relations online onbulletin board systems, interdisciplinary research methods are used combining thetheories of communication, sociology and computer science. Recognition of publicrelations online is divided into two aspects-recognition of events which aremanipulated by public relations online and recognition of internet users who controlthe developing process of the network events. The main work of the study:First, based on Lasswell’s Model, manipulation model of network events isproposed. According to the model and characteristics of bulletin board systems, anindex system is set up to judge whether a network event is controlled by publicrelations online and indexes in the system are proved to be valid.Second, recognition of events is analyzed and verified as a classification problem.The verification of an actual case proves that the event recognition model built bySupport Vector Machine (SVM) is acceptable for identification of events with theaccuracy77.9%. Third, recognition of internet users is studied as a feature selection problem andan effective evaluation function is constructed. Using the evaluation function, thosewho hype the actual internet event studied are recognized.The result of this study can be applied to provide a reference to judge theauthenticity of network events, supply evidence for the governments to take measuresunder certain circumstances and offer a train of thought for information securitycompanies to produce new products.

Related Dissertations
More Dissertations