DocumentCode :
2422576
Title :
Predictive Model for Internet Public Opinion
Author :
Zeng, Jianping ; Zhang, Shiyong ; Wu, Chengrong ; Xie, Jianfeng
Author_Institution :
Fudan Univ., Shanghai
Volume :
3
fYear :
2007
fDate :
24-27 Aug. 2007
Firstpage :
7
Lastpage :
11
Abstract :
The Internet is becoming a spreading platform for the public opinion. It´s important to model the Internet public opinion activity as accurate as possible. Hidden Markov Model (HMM) is introduced to describe the activity of the Internet public opinion. The state of the Internet public opinion is represented as the hidden state of HMM, and its characteristic is considered as the visible states in HMM. Then the HMM is used to predict the characteristic changing of the Internet public opinion using two different algorithms under the circumstances of different amount of characteristic data gathered. Experiments on the Internet public opinion are done and the result shows that the modeling method and prediction algorithms are effective.
Keywords :
Internet; hidden Markov models; Internet public opinion activity; hidden Markov model; prediction algorithm; predictive model; Hidden Markov models; IP networks; Information technology; Internet; Mathematical model; Mathematics; Performance analysis; Prediction algorithms; Predictive models; Stability;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Fuzzy Systems and Knowledge Discovery, 2007. FSKD 2007. Fourth International Conference on
Conference_Location :
Haikou
Print_ISBN :
978-0-7695-2874-8
Type :
conf
DOI :
10.1109/FSKD.2007.453
Filename :
4406191
Link To Document :
بازگشت