Dissertation > Excellent graduate degree dissertation topics show

Study on Crisis Early-warning of Internet Public Opinion Based on BP Neural Network

Author: XueQuanQuan
Tutor: LeZhongJian
School: Jiangxi University of Finance
Course: Management Science and Engineering
Keywords: Internet Public Opinion Crisis Early-warning Index System Back-propagation Neural Network
CLC: F224
Type: Master's thesis
Year: 2010
Downloads: 479
Quote: 1
Read: Download Dissertation

Abstract

With the rapid development of Internet, network is called "the fourth media" following newspaper, the radio and TV. It has become one of the main carriers of reflecting public opinion. The freedom and diversity of Internet public opinion is a "two-sided sword". It is an ideal window of the lower levels on the up and is an effective way of the government knowing public opinion rapidly and directly. But it is used by some people to produce false information and spread rumor, so that it brings about some threat on social stability and safety. Recent years, significant events of Public opinion are greatly increasing via Internet in China. It arouses our country’s high attention. The demand of early-warning about Internet public opinion is urgent. Establishing a crisis early-warning system of Internet public opinion can help the relevant departments detect central issues of society, the negative information or signs of potential problems timely and make early warning, ensuring the social safety and stability and promoting national harmony.This paper develops on the theoretical knowledge about Internet public opinion and crisis early-warning. The core content is to establish the model of crisis early-warning of Internet public opinion based on BP neural network, including two parts:the first part of the model is to establish the index system of crisis early-warning of Internet public opinion from 3 levels,4 aspects,11 specific indicators, and it describes the 11 specific indicators qualitatively to make the judgment at Internet public opinion more scientific and objective. The index system of crisis early-warning of Internet public opinion is an important defense line of supervision, crisis analysis and early warning of Internet public opinion. The second part is to establish the model. The paper establishes the model of crisis early-warning of Internet public opinion based on BP neural network using the neural network toolbox of Matlab7.0. Because of the complexity, non-deterministic and nonlinear in the process of crisis early-warning of Internet public opinion, the paper applies BP neural network in crisis early-warning of Internet public opinion. And it demonstrates the effectiveness of this model by the event of entrapment in shanghai. Finally, the paper provides some strategies and management suggestions for dealing with the crisis of Internet public opinion. The main results are as follows:(1) Establishing a new crisis early-warning index system of Internet public opinion, enriching the discovery on the index of crisis early-warning of Internet public opinion.(2) Presenting practical quantitative methods on the basis of the qualitative description about the index of crisis early-warning of Internet public opinion, to make the index system proposed in the text more scientific.(3) Establishing the model of crisis early-warning of Internet public opinion based on BP neural network, and demonstrating the feasibility and effectiveness of this model by the research on the specific case.Internet public opinion casts light on public opinion timely, concerned and valued by all levels of society. This paper provides references for the index research and the early-warning method of crisis early-warning of Internet public opinion, and provides some Suggestions for dealing with the crisis of Internet public opinion.

Related Dissertations

CLC: > Economic > Economic planning and management > Economic calculation, economic and mathematical methods > Economic and mathematical methods
© 2012 www.DissertationTopic.Net Mobile