Abstract
ABSTRACT A classified prediction model for Internet public opinions (IPO) with data mining (DM) technology based on the theory of fast dual-cycle level set is proposed in this paper to improve IPO prediction precision. Firstly, a feature extraction algorithm is applied to model the weighting of IPO feature values, which reflect its significance in IPO. Then IPO similarity is used to present the degree of similarity between two IPO topics with respect to semantic meaning. At last, a certain hot topic on the Internet is used for simulation test on the model proposed. The results show that the model proposed herein can accurately predict the trends of IPO changes and improve the IPO prediction precision.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
