SENTOSO, SATRIYO BAGAS (2025) ANALISA SENTIMEN PADA FENOMENA FUFUFAFA PADA APLIKASI X DENGAN ALGORTIMA NAIVE BAYES. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
01 COVER.pdf Download (485kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
02 BAB 1.pdf Restricted to Registered users only Download (110kB) |
|
|
Text (BAB II)
03 BAB 2.pdf Restricted to Registered users only Download (344kB) |
|
|
Text (BAB III)
04 BAB 3.pdf Restricted to Registered users only Download (240kB) |
|
|
Text (BAB IV)
05 BAB 4.pdf Restricted to Registered users only Download (1MB) |
|
|
Text (BAB V)
06 BAB 5.pdf Restricted to Registered users only Download (42kB) |
|
|
Text (DAFTAR PUSTAKA)
07 DAFTAR PUSTAKA.pdf Restricted to Registered users only Download (106kB) |
|
|
Text (LAMPIRAN)
08 LAMPIRAN.pdf Restricted to Registered users only Download (697kB) |
Abstract
This study aims to analyze public sentiment and emotional responses to the "FufuFafa" phenomenon that went viral on Application X (formerly Twitter) using the Naïve Bayes Classifier (NBC) method. The data was collected through web scraping techniques using the Twitter API, resulting in 3,835 tweets containing the keywords "FufuFafa" and "Gibran" from September to October 2024. The analysis includes sentiment classification (positive, negative, neutral) and emotional responses (anger, sadness, fear, happiness, surprise). The data processing involved text mining stages such as cleaning, tokenization, normalization, stemming, and stopword removal. Sentiment classification was performed using three Naïve Bayes algorithms (Gaussian, Multinomial, Bernoulli). Gaussian Naïve Bayes achieved the highest accuracy for sentiment analysis on test data (83%), followed by Multinomial (81%) and Bernoulli (73%). For emotional response analysis, Gaussian and Multinomial achieved 98% accuracy, while Bernoulli only reached 59%. The analysis of emotional responses revealed that most tweets were labeled as NaN or unidentifiable (3,266 tweets, 85.77%), followed by anger (263 tweets, 6.91%), fear (116 tweets, 3.05%), happiness (65 tweets, 1.71%), surprise (53 tweets, 1.39%), and sadness (45 tweets, 1.18%). These results indicate that the majority of tweets did not exhibit clear emotional responses. However, among the tweets that were successfully identified, the dominance of anger and sadness suggests significant dissatisfaction and disappointment regarding the issue, providing key insights into the dynamics of public opinion on social media. This study offers valuable insights into public sentiment and emotional dynamics in social media discussions surrounding sensitive political issues, while also providing critical information for policymakers and future researchers. Keywords : Sentiment Analysis, Naïve Bayes, FufuFafa, Emotional Response, Social Media Penelitian ini bertujuan untuk menganalisis sentimen publik dan respons emosional terhadap fenomena "FufuFafa" yang viral di Aplikasi X (sebelumnya Twitter) dengan menggunakan metode Naïve Bayes Classifier (NBC). Data diperoleh melalui teknik web scraping menggunakan API Twitter, menghasilkan 3.835 tweet yang mengandung kata kunci "FufuFafa" dan "Gibran" selama periode September hingga Oktober 2024. Analisis mencakup klasifikasi sentimen (positif, negatif, netral) dan respons emosional (marah, sedih, takut, senang, terkejut).Data yang diproses melewati tahapan text mining seperti pembersihan, tokenisasi, normalisasi, stemming, dan penghapusan kata umum (stopword removal). Klasifikasi sentimen dilakukan menggunakan tiga algoritma Naïve Bayes (Gaussian, Multinomial, Bernoulli). Gaussian Naïve Bayes mencapai akurasi tertinggi untuk analisis sentimen pada data uji (83%), diikuti oleh Multinomial (81%) dan Bernoulli (73%). Untuk analisis respons emosi, Gaussian dan Multinomial mencapai akurasi 98%, sementara Bernoulli hanya mencapai 59%. Hasil analisis respons emosi menunjukkan sebagian besar tweet bersifat NaN atau tidak teridentifikasi (3.266 tweet, 85,77%), diikuti oleh emosi marah (263 tweet, 6,91%), takut (116 tweet, 3,05%), senang (65 tweet, 1,71%), terkejut (53 tweet, 1,39%), dan sedih (45 tweet, 1,18%). Hasil ini menunjukkan mayoritas tweet tidak memberikan respons emosional yang jelas. Namun, di antara tweet yang berhasil diidentifikasi, respons masyarakat umum pada Aplikasi X yang didominasi oleh emosi marah dan sedih menunjukkan adanya kekecewaan serta ketidakpuasan terhadap isu yang berkembang, yang menjadi indikasi penting dinamika opini publik di media sosial.Penelitian ini memberikan wawasan tentang sentimen publik dan dinamika emosional dalam diskusi media sosial mengenai isu politik yang sensitif, serta menawarkan informasi yang berharga bagi pembuat kebijakan dan peneliti di masa mendatang. Kata kunci: Analisis Sentimen, Naïve Bayes, FufuFafa, Respon Emosi, Media Sosial
Actions (login required)
 |
View Item |
