Twitter Sentiment Analysis Classification to Assess Public Opinion on Football Matches Using the Naïve Bayes Method

Yuli Astuti (Program Studi Informatika Universitas Amikom Yogyakarta Yogyakarta, Indonesia)
Hafiidh Khoiru Pradana (Program Studi Informatika Universitas Amikom Yogyakarta Yogyakarta, Indonesia)
Dewi Anisa Istiqomah (Program Studi Informatika Universitas Amikom Yogyakarta Yogyakarta, Indonesia)
supriatin supriatin (Program Studi Informatika Universitas Amikom Yogyakarta Yogyakarta, Indonesia)
Ninik Tri Hartati (Program Studi Informatika Universitas Amikom Yogyakarta Yogyakarta, Indonesia)

Journal of Intelligent Software Systems
Vol 2, No 2 (2023): December 2023

 The Kanjuruhan tragedy has attracted many comments on various social media platforms. This research will compare the number of positive and negative comments on Twitter and social media and determine the accuracy of the classification method used. The data used in this study consisted of 2052 pieces, consisting of 1015 positive and 1037 negative pieces. To determine the effect of the amount of training data on the resulting accuracy, testing will be carried out three times with different combinations of training data and test data, namely 70:30, 80:20, and 90:10. The results of this study obtained the highest accuracy value of 79.6%. This program can be developed for other social media platforms such as Facebook, Instagram, and others