Forecasting with Python, case study: visits to social networks in Ecuador with machine learning

Abstract

This article emphasizes the importance of respect in the use of social networks in Ecuadorian society, highlighting their diversity and the need to foster constructive dialogue rather than harmful online confrontation. A methodology employing machine learning techniques and statistical modeling, such as the ARIMA model, is provided to predict web traffic on social networks in Ecuador. In addition, various backtesting strategies are discussed to evaluate and improve the accuracy of the model over time. The results indicate significant growth in the number of social network users in Ecuador, with a  focus on the ARIMA model as effective for time series prediction, although exploration of additional approaches and continued improvements are suggested in future research. This study contributes to a better understanding of the impact of social networks in Ecuadorian society and provides a methodological basis for forecasting web traffic on these platforms in the future.