E-LEARNING ECONOMIC CRISIS

Abstract

The potency of e-learning varies from context to context, however, has also been shown to become a valuable method of predicting an economic crisis. Basically, e-learning becomes an accessible and efficient method to avoid financial and social blockages worldwide. Due the popularity of deep learning for analysing text data, with really promising results, for instance, in financial forecasting from news focused on specific topics, a new vision of predicting an economic crisis may be the foundation of an intelligent economic theory. This study explores machine learning algorithms to predict different economic stressful events. Also, it addresses the problem of correlating the open information provided by economic publications and social media with impact on the economic behaviour. The goal of this paper is to implement an e-learning system able to provide a set of valid information about a potential economic crisis, using a dataset of financial and economic topics. Consequently, there is a need to evaluate the performance of e-learning in the domain of financial economics as part of ongoing quality of life improvements efforts. The results can be seen as a starting point for broader research in the same field. The entire research was based on history of previous economic crisis and the entire chain of events extracted from the dataset consisting of economic news. Based on these results, using a newspaper collection chronologically ordered from 2008 to 2018, with an error margin of approximately one or two years, the signs of the next economic crisis can be already observed.