Twitter as a predictive system: A systematic literature review
通过系统性文献综述，评估和分类推特用户生成内容在预测应用中的潜力，识别了社交网络分析和公共卫生等领域的空白与机会，对学者和商业领袖有参考价值。
Millions of people use Twitter daily, posting thousands of messages and interacting with their peers. This research aims to evaluate and classify the predictive potential of the Twitter social platform through the intelligent analysis of user-generated public big data analytics. A systematic literature review (SLR) covering Web of Science, IEEE, Scopus and other databases identified the gaps and opportunities for developing predictive applications of User-Generated Content (UGC) on Twitter since 2006. Our research is a practical contribution to the use of Twitter data as a predictive system. A wide variety of application domains, highlighting social network analysis and public health, have been identified by applying innovative techniques for conducting a massive SLR, leveraging machine learning and graph analysis. The results give rise to new research lines with implications for both scholars and business leaders.