Internet public opinion event automatic early warning system based on event propagation characteristics

Abstract

The invention relates to an internet public opinion event automatic early warning system based on event propagation characteristics, which comprises a data acquisition module, a data preprocessing module, an event forming module, a characteristic capturing module and a decision making module which are connected in sequence, the data acquisition module obtains news message data of all media platforms of the whole network through a crawler system, carries out primary processing and outputs a basic data result to the data preprocessing module, and the data preprocessing module cleans disordered data in various formats from the whole network. The event forming module obtains information from the data preprocessing module, carries out event forming processing, carries out feature collection on a formed event and outputs features describing event attributes to the decision making module; and the decision-making module carries out decision-making processing according to the characteristics describing the event attributes and finds the most suitable decision-making module parameters through multiple experiments, so that the automatic public opinion event early warning function is completed in real time with the maximum accuracy.

Description

technical field [0001] The invention relates to a technology in the field of information processing, in particular to an automatic early warning system for Internet public opinion events based on event propagation characteristics. Background technique [0002] As the network develops faster, the information becomes more transparent. In recent years, there have been various public security incidents at various levels. Due to the particularity of these incidents, they have attracted enormous attention in the online society, and the erroneous remarks of some media reporters have caused the anger of netizens. According to the development and evolution law of online public opinion on emergencies in the past, some potential characteristics that will form major public opinion events can be pre-learned from it, so as to help the public opinion monitoring system to more accurately identify dangerous public opinion events, give early warnings, and achieve preventive results. . For t...

AI Technical Summary

Problems solved by technology

[0003] By consulting the literature, it is found that the main prediction and early warning methods of the existing network public opinion are based on statistical learning systems, Bayesian networks, gray systems and traditional methods based on fuzzy theory to make model predictions, and they are all based on text Particles, which have a great loss on the performance of the system, because the events described in various texts are different, so this will cause problems in the prediction and early warning of events

Images