Clasificación de publicaciones en redes sociales mediante el uso de técnicas de aprendizaje automático

Abstract

Abstract: Through machine learning, high-level classifications and predictions can be established. However, this tool has not yet been widely used on various digital platforms, such as social media, for information classification. The objective of this project was to classify social media posts using machine learning techniques. The methodology was applied and descriptive, analyzing sentiments in posts through Bayesian models, Support Vector Machine (SVM), and Random Forest. Results showed that all three models provided accuracy above 0.74, indicating a high classification capacity. The SVM model yielded the best results with the given data, while the Bayesian network model was the fastest in prediction time.