Big data driven election public opinion prediction method

Abstract

The invention discloses a big data driven election public opinion prediction method, and belongs to the field of data mining. The method includes the specific steps that firstly, big data information sources are picked out according an Internet network use report and a region website ranking of an election country or region; then, each type of information sources are analyzed, and public opinion prediction indexes are established; then, extracted multisource prediction indexes are integrated, and a current Internet public approval rate is formed; then, opinion poll reports are collected, the approval rates of people of all ages in the reports on candidates are integrated, and an offline opinion poll approval rate value is formed; in consideration of the population structure and composition and the net citizen age distribution of the election country or region, candidate Internet approval rate values and the offline opinion poll approval rate value are integrated, the candidate approval rate value at a next time node is predicted through a moving average method, and the prediction result is updated at the daily frequency. The method has the advantages of wide data sources, the short prediction period, high real-time performance and the like, and has great application value in the fields of public opinion monitoring, viewpoint analysis and the like.

Description

technical field [0001] The invention belongs to the field of data mining and relates to prediction technology, in particular to a big data-driven election public opinion prediction method. Background technique [0002] With the development of network technology, the group of netizens continues to expand. According to the "Measuring the Development of the Information Society Report" released by the International Telecommunication Union (ITU), as of 2015, the number of global Internet users has reached 3.2 billion, accounting for 43.4% of the global population. The Internet usage rate in Europe is as high as 82.1%, followed by the United States with an Internet usage rate of 60%. It can be seen that the Internet has gradually integrated into people's lives, especially in developed countries. With the vigorous development of search engines, portal websites, and social networks, multi-source network platforms not only provide people with rich information, but also provide spac...

AI Technical Summary

Problems solved by technology

However, the existing big data-based election prediction methods mainly rely on a single channel of information sources, and the method has weak transferability; and because the main user groups of the Internet are mainly young people, the age distribution of network data is biased.

Images