Football Supporters Club Opinion Analysis using Recurrent Neural Network

In this study, we conducted a sentiment analysis of the opinions of football club supporters using the Recurrent Neural Network (RNN) algorithm with Long Short-Term Memory (LSTM). We adopted the best model for each club to predict the sentiments expressed on social media Twitter. The experimental results show that using the RNN algorithm with LSTM produces excellent accuracy in analyzing the opinions of football club supporters. In the Arsenal dataset, accuracy reaches 80% with a training data sharing scheme of 90% and 10% testing. The accuracy of the Liverpool club dataset reaches 82% with the same data-sharing scheme. Meanwhile, on the Manchester United club dataset, this algorithm also achieves 80% accuracy with a similar data-sharing scheme. These findings indicate that applying the RNN modeling technique with LSTM is very effective in analyzing the opinions of football club supporters on social media platforms. These sentiment results can provide valuable insights for these clubs to better understand their supporters’ feelings and views.