A method of public opinion quantification based on neural network

Abstract

The invention discloses a method for quantifying public opinion based on a neural network. The steps are: 1) using crawler technology to collect user's public opinion text data on a social platform, and storing the collected data in a database in the form of text; Perform data preprocessing on the public opinion text data; 3) According to the preprocessed data, construct a public opinion analysis data set; 4) Build a network model, and use the public opinion text and associated factor data in the public opinion analysis data set to train the network model; 5) The trained network model is used to quantify the public discourse text; this method can quantify the symbolic text by means of neural network, and uses the weight calculation of multi-feature fragments to determine the influence weight of various factors. Compared with the traditional method, this method is more flexible and more robust. From a social and economic point of view, this method can grasp the hotspots, follow the increase of the amount of data, become more objective and rational, and be more in line with the actual results of public opinion.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence and deep learning, in particular to a public opinion quantification method based on a neural network. Background technique [0002] With the vigorous development of the Internet, the number of users increases rapidly, and the information generated by users online continues to grow, which provides a huge amount of data for the research and development of big data. Appearing on the Internet, it reflects the subjective understanding and value of society from various social media. How to extract some business risk information from it and quantify it into a digital form is of great value. [0003] At present, there is no mature risk quantification method in the prior art. The difficulty lies in the fact that the symbolic text cannot be digitized in an objective form. There are many factors that need to be considered for the public opinion text. The weight distribution cannot be object...

AI Technical Summary

Problems solved by technology

[0003] At present, there is no mature risk quantification method in the existing technology. The difficulty is that it is impossible to digitize symbolic text in an objective form. For public opinion texts, there are many factors that need to be considered. This kind of weight distribution cannot be objectively judged statistically.

Images