基于机器学习的网络舆情预测研究综述

Literature review of online public opinion prediction using machine learning

大数据时代来临,社交网络和自媒体平台为公众提供了解多元文化的渠道。与此同时,一些负面、虚假或未被证实的消息同样可以得到传播。因此,在舆情形成初期,实时监看和预测未来的舆情走势成为重中之重。传统的网络舆情监看依靠人工信息汇集和分析预测舆情走势,随着机器学习技术的发展,基于机器学习的预测方法已在各个行业取得成就。文中对应用于网络舆情预测的主要机器学习模型进行梳理和理论介绍。此外,对当前应用机器学习于网络舆情预测的主要方法进行全面综述、对引入的关键技术进行整理介绍并探讨未来发展方向。

With the advent of the era of big data, social networks and media platforms provide citizens with a channel to understand diverse cultures. However, some negative, false and unverified information can also be widely spread. Therefore, in the early stage of public opinion, real-time monitoring and prediction of the future trend have become a top priority. Traditional network public opinion monitoring relies on artificial information collection to predict the trend of public opinion. With the development of machine learning technology, prediction research based on machine learning methods has made some achievements in various industries. In this paper, the main machine learning models used for online public opinion prediction are sorted out and introduced theoretically. In addition, the popular methods of the current application of machine learning in network public opinion prediction are comprehensively summarized, the key technologies introduced are sorted out and introduced, and the future development direction of the methods is discussed.