Skripsi
SENTIMEN ANALISIS KOMENTAR INSTAGRAM PADA AKUN MATA NAJWA MENGGUNAKAN METODE SELEKSI FITUR INFORMATION GAIN DAN CHI-SQUARE DENGAN RANDOM FORET
Instagram is one of the most popular social media platforms that is not only used to share pictures and videos, but also as a means of discussion on various social, political and cultural issues. One of the most active and influential accounts in Indonesia is the Mata Najwa account, which often triggers public discussion through the comments section. The large number of incoming comments creates challenges in managing and analyzing the data, especially due to the presence of comments that are ambiguous, irrelevant, or even contain hate speech. This research aims to develop software for classifying Instagram comments on the Mata Najwa account using the Random Forest algorithm. The dataset used consists of 574 comments categorized into three classes: positive, negative, and neutral. Three testing approaches were conducted: Random Forest without feature selection, Random Forest with feature selection using Information Gain, and Random Forest with feature selection using Chi-Square. The test results show that the Random Forest method without feature selection produces an average accuracy of 0.7536, precision of 0.4606, recall of 0.4163, and F-Measure of 0.4164. Meanwhile, the Random Forest method with Information Gain produced the highest F-Measure of 0.4567, and the method with Chi-Square recorded the highest accuracy of 0.7565. This research shows that the Random Forest algorithm can be effectively used in the classification of social media comments.
| Inventory Code | Barcode | Call Number | Location | Status |
|---|---|---|---|---|
| 2507004181 | T178886 | T1788862025 | Central Library (REFERENCE) | Available but not for loan - Not for Loan |
No other version available