Analisis sentimen pengguna Twitter terhadap pogram Pembatasan Sosial Berskala Besar (PSBB) transisi di Jakarta menggunakan metode naive bayes dan decision tree

Yudhistira, Afrizandy Bayu (2022) Analisis sentimen pengguna Twitter terhadap pogram Pembatasan Sosial Berskala Besar (PSBB) transisi di Jakarta menggunakan metode naive bayes dan decision tree. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 10611710000093-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (1MB)

Abstract

PSBB Transisi pertama kali diterapkan di DKI Jakarta pada bulan Juni 2020, yang merupakan kebijakan lanjutan setelah adanya PSBB yang dinilai masih kurang efektif. Kebijakan tersebut membuat banyak pengguna Twitter menyuarakan opini mereka. Opini tersebut mengandung kalimat positif dan negatif yang dapat dimanfaatkan untuk melakukan analisis sentimen dengan menggunakan metode Naïve Bayes dan Decision Tree. Kedua metode tersebut nantinya akan dibandingkan untuk mengetahui metode mana yang paling baik digunakan dalam klasifikasi sentimen berdasarkan nilai accuracy dan G-Mean. Hasil yang didapatkan menunjukkan bahwa pada analisis sentimen, opini pengguna Twitter terhadap program PSBB Transisi terdiri dari 70% tweet yang mengandung sentimen negatif dan 30% tweet yang mengandung sentimen positif. Performa algoritma Decision Tree lebih baik dalam mengklasifikasikan data ulasan pengguna Twitter dibandingkan dengan algoritma Naïve Bayes. Hal tersebut ditunjukkan dari nilai accuracy algoritma Decision Tree sebesar 76%, yang lebih tinggi dibandingkan dengan nilai accuracy algoritma Naïve Bayes sebesar 73%, serta nilai G-Mean algoritma Decision Tree sebesar 62%, yang lebih tinggi dibandingkan dengan nilai G-Mean algoritma Naïve Bayes sebesar 24%.
==============================================================================================================================
The transition to Large-Scale Social Restrictions (PSBB) was first implemented in DKI Jakarta in June 2020 as a follow-up policy to the previous PSBB, which was considered less effective. This policy led many Twitter users to express their opinions. These opinions contained both positive and negative sentiments, which could be utilized for sentiment analysis using the Naïve Bayes and Decision Tree methods. The two methods were then compared to determine which method was more effective for sentiment classification based on accuracy and G-Mean values. The results of the sentiment analysis showed that 70% of Twitter users’ opinions regarding the PSBB Transition program contained negative sentiments, while 30% contained positive sentiments. The Decision Tree algorithm demonstrated better performance in classifying Twitter user review data than the Naïve Bayes algorithm. This was indicated by the Decision Tree algorithm’s accuracy of 76%, which was higher than the Naïve Bayes algorithm’s accuracy of 73%. Furthermore, the Decision Tree algorithm achieved a G-Mean value of 62%, which was higher than the Naïve Bayes algorithm’s G-Mean value of 24%.

Item Type:	Thesis (Other)
Additional Information:	519.53 Yud a-1
Uncontrolled Keywords:	Accuracy, Analisis Sentimen, Decision Tree, G-Mean, Naïve Bayes, Klasifikasi, PSBB Transisi, Accuracy, Classification, Decision Tree, G-Mean, Large-scale social restrictions the transition, Naïve Bayes Sentiment Analysis
Subjects:	Q Science > QA Mathematics > QA278.55 Cluster analysis
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	magang .
Date Deposited:	05 Oct 2026 03:41
Last Modified:	05 Oct 2026 03:41
URI:	http://repository.its.ac.id/id/eprint/145243

Actions (login required)

View Item