Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : software development digital business intelligence and computer engineering

PENERAPAN SMOTE PADA ANALISIS SENTIMEN OPINI PUBLIK TERHADAP PROGRAM MAKAN SIANG GRATIS PADA MEDIA SOSIAL X MENGGUNAKAN ALGORITMA NAÏVE BAYES Kurniawan, Widya; Indriani, Nur Salsabila Citra; Utami, Amelia
Pengembangan Perangkat Lunak, Kecerdasan Bisnis Digital, dan Teknik Komputer Vol 5 No 1 (2026): SESSION (SEPTEMBER)
Publisher : Politeknik Negeri Banyuwangi Jl. Raya Jember km. 13 Labanasem, Kabat, Banyuwangi, Jawa Timur (68461) Telp. (0333) 636780

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57203/session.v5i1.2026.7-13

Abstract

The Free Lunch Program is a government policy to improve public welfare. Evaluation of the effectiveness of this program is important to ensure that the benefits are as expected. This study analyzes public opinion about the Free Lunch Program through social media X with the sentiment analysis method using the Naïve Bayes algorithm. Data was obtained using the web crawling technique using the keyword "free lunch," resulting in 2,104 tweets that were processed through the preprocessing stage, including text cleaning and removing stopwords. Weighting was done using the TF-IDF method. The model was developed with Multinomial Naïve Bayes, using 80% of the data for training and 20% for testing and applying the SMOTE technique for data imbalance. The results showed an accuracy of 92% and challenges in identifying neutral sentiment. It is expected that the application of SMOTE to sentiment analysis with Naïve Bayes can improve the accuracy of imbalanced data classification, resulting in a more objective and in-depth analysis.