Random forest na análise de sentimentos em jogos digitais

Abstract

This study describes the application of the Random Forest algorithm in sentiment analysis of digital games, categorizing text into positive, negative, or neutral sentiments. The research aims to deepen the understanding of player interactions with games, providing relevant insights for companies in a constantly evolving industry. The methodology employed includes the collection of a dataset containing 640,987 player reviews and meticulous preprocessing to ensure data quality. The dataset was divided into 80% for training (512,790 reviews) and 20% for testing (128,197 reviews), with a sample of 109,075 reviews analyzed. The Random Forest model achieved an accuracy of 89%, a precision of 99% for positive sentiments, and 63% for negative sentiments, and a recall of 88% for positive sentiments and 94% for negative sentiments. The confusion matrix results indicate an overall accuracy of 89%. These metrics demonstrate the model's effectiveness in accurately classifying player sentiments. In summary, this study provides valuable data driven insights into player feedback, offering a practical tool for companies to enhance their products and strategies in the competitive gaming market.