System for automatic identification, classification and reporting of cross-domain public opinions.

Abstract

Provided is a system for automatic identification, classification and reporting of cross-domain public opinions. With popularization of smartphone and development of WeChat, microblogs and other internet technologies, the information dissemination speed is remarkably improved. Some petty things which can be handled duly cannot be found and solved and continuously fermented and finally bring severe influences such as the 'Shanghai treading' accident and the 'Qingdao prawn' event; meanwhile, good suggestions and supervision of some citizens cannot be also submitted to relevant functional departments in time, for example, the pyramid sale behavior existing in a certain housing estate and issue through microblog, supervision strengthening, loss of well lids in a certain road section and supplementation. For solving the problems, the system comprises the following steps that 1, mass data is searched through fixed-point Wi-Fi, WeChat positioning and other technologies; 2, the public opinions are obtained through big data analysis; 3, the public opinions are automatically classified and reported to the corresponding functional departments. The malignant events can be found and avoided to some degree in time, the opinions of citizens can be also comprehensively searched in time, governmental service can be improved, and finally the life quality of the citizens in a smart city can be improved.

Description

technical field [0001] The present invention belongs to the field of big data, and more precisely, relates to Internet public opinion monitoring technology, mainly text analysis technology. Background technique [0002] The emergence of the Internet has accelerated the dissemination of information and facilitated communication between people. But at the same time, crimes through the Internet are also a problem, such as sensitive political activities, online economic fraud, etc. Therefore, the public security department was the first to monitor public opinion. Early technologies generally filter by keywords, sensitive words, etc. Although this method is simple, it is very effective. With the development of the times, some criminals have gradually mastered some anti-reconnaissance methods, such as splitting sensitive words and other tricks. At this time, the method of filtering through keywords is difficult to work. In order to adapt to this change, the current public opinio...

AI Technical Summary

Problems solved by technology

[0003] 1) The current public opinion discovery is mostly based on the analysis of long texts such as forums and web pages, and the public opinion analysis of short texts such as Weibo and WeChat is not enough

[0004] 2) Most of the current public opinion discovery is based on a single field, and there is no technology for automatic analysis, classification and reporting of cross-field public opinion.

Images