Predictive Sentiment Analysis Model Regarding the Variation of the Dollar Exchange Rate

Bernabe-Polo J.; Guzmán-Monteza Y.; Puelles-Bulnes E. · Communications in Computer and Information Science (2023) · DOI: 10.1007/978-3-031-35445-8_9

Resumen

The benefits of sentiment analysis have become a topic of enormous interest in recent years. Obtaining information about this massive bank of social networks is essential. The dollar exchange rate for Peru, as well as for any country, is one of the most important economic indicators. Investment decisions made by national or international companies, as well as by governments, are based on the exchange rate. Exchange rate prediction is essential for future investors and companies. Twitter is the frequent mean by which people indicate their points of view about a specific topic. Therefore, in the present investigation, the proposal is described to analyze the trend of Twitter users for the dollar variation using open banking data and data from social networks as pillars. In the first phase, NB and RF were used and obtained an accuracy of 90% for both models’ four (04) combinations. For the results of the second phase of the research, better results were obtained with the SVM algorithm, achieving a percentage of accuracy of 92.82% to predict the variation of the dollar exchange rate.

Instituciones: Universidad Nacional Mayor de San Marcos