Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/165816| Title: | Penerapan Metode Naïve Bayes Classifier terhadap Sentimen Masyarakat pada Tagar #Kaburajadulu di Media Sosial X |
| Other Titles: | Application of the Naïve Bayes Classifier Method to Public Sentiment on the #Kaburajadulu Hashtags on X Social Media |
| Authors: | Suwanda, Bayu Suriaatmaja Purnama, Eni |
| Issue Date: | 2025 |
| Publisher: | IPB University |
| Abstract: | Pada awal Februari 2025 masyarakat Indonesia ramai memperbincangkan tagar #Kaburajadulu di media sosial X. Interaksi melalui tagar membentuk suatu jaringan komunikasi dan sentimen masyarakat. Penelitian ini bertujuan untuk melihat bagaimana penerapan metode Naïve Bayes Classifier terhadap sentimen masyarakat dan menguji nilai akurasinya serta bentuk komunikasi yang terbentuk. Metode penelitian yang digunakan adalah deskriptif kuantitatif dengan menggunakan purposive sampling. Sampel dalam penelitian ini sejumlah 100 responden yang diambil melalui rumus Lemeshow. Data yang diperoleh diproses menggunakan Python pada Google Colab dan diperkuat dengan SPSS versi 26 untuk menguji hipotesis penelitian menggunakan regresi logistik multinominal. Berdasarkan analisis diperoleh hasil bahwa metode Naïve Bayes Classifier berhasil menganalisis sejumlah 910 tweet dari 1000 tweet dengan nilai akurasi sebesar 56%. Diperkuat hasil uji regresi logistik multinominal yang membentuk komunikasi Interpersonal dan Komunikasi Massa terhadap cara pengguna X menyampaikan pesan yang menghasilkan sentimen. In early February 2025, the Indonesian people are busy discussing the hashtag #Kaburajadulu on social media X. Interaction through hashtags forms a network of communication and community sentiment. This study aims to see how the Naïve Bayes Classifier method is applied to public sentiment and test its accuracy value and the form of communication formed. The research method used was quantitative descriptive using purposive sampling. The sample in this study was 100 respondents taken through the Lemeshow formula. The data obtained was processed using Python on Google Colab and reinforced with SPSS version 26 to test the research hypothesis using multinominal logistic regression. Based on the analysis, the results were obtained that the Naïve Bayes Classifier method managed to analyze a total of 910 tweets out of 1000 tweets with an accuracy value of 56%. It was strengthened by the results of the multinominal logistics regression test that formed Interpersonal Communication and Mass Communication on the way X users conveyed messages that generated sentiment. |
| URI: | http://repository.ipb.ac.id/handle/123456789/165816 |
| Appears in Collections: | UT - Digital Communication and Media |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| cover_J0301211375_367b38275bd542f48b01f80299f63b09.pdf | Cover | 251.54 kB | Adobe PDF | View/Open |
| fulltext_J0301211375_61192ed87e49411ba2359c36048bc830.pdf Restricted Access | Fulltext | 879.65 kB | Adobe PDF | View/Open |
| lampiran_J0301211375_134b19065b774f3d805571ebf726af8b.pdf Restricted Access | Lampiran | 456.85 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.