Lahdji, Muhammad Rayhan (2022) Klasifikasi Sentimen pada Ulasan Pengguna Aplikasi PeduliLindungi di Google Play Store Menggunakan Metode Naïve Bayes, Regresi Logistik, dan Regresi Logistik dengan Regularisasi L-1. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06211740000075-Undergraduate_Thesis.pdf - Accepted Version Download (1MB) | Preview |
Abstract
COVID-19 adalah penyakit infeksi pernapasan yang menyebar ke seluruh dunia dan mengakibatkan pandemi. Dampak pandemi amat parah baik dilihat dari segi ekonomi, sosial, kesehatan fisik, bahkan kesehatan mental. Pemerintah Indonesia berkewajiban menjaga keselamatan rakyat Indonesia, dan termasuk dalam kewajiban tersebut adalah melaksanakan mitigasi dampak dan penanganan COVID-19. Strategi yang digunakan antara lain adalah dengan pengawasan kesehatan (health surveillance) dengan aplikasi mobile yakni PeduliLindungi. Penting untuk mengetahui sentimen masyarakat terhadap aplikasi tersebut agar penggunaannya dapat lebih luas dan efektif. Penelitian ini menggunakan metode klasifikasi sentimen Naïve Bayes, Regresi Logistik, dan Regresi Logistik dengan Regularisasi L-1. Didapat bahwa model yang paling baik dinilai dari metrik accuracy, precision, recall, dan F1-score adalah regresi logistik dengan regularisasi L-1.
================================================================================================
COVID-19 is an infectious respiratory disease that causes a worldwide pandemic. Its economic, social, physical health, and mental health impacts on Indonesian society is readily perceivable and noticeable. The government of Indonesia is responsible to implement necessary measures in order to handle and mitigate the pandemic’s effects to Indonesian citizens. Such measures include the use of health surveillance application PeduliLindungi. It is crucial that user feedback and sentiment is analyzed in order to broaden the application’s usage and maximize its effectiveness. This research uses naïve Bayes, logistic regression, and L1-regularized logistic regression to classify user review sentiments. This research concludes that L1-regularized logistic regression outperforms other models using accuracy, precision, recall, and F1-score as performance metrics.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Sentimen, Ulasan, Naive Bayes, Regresi Logistik, Regularisasi L-1, Sentiment, Review, Logistic Regression, L-1 Regularization |
| Subjects: | Q Science Q Science > QA Mathematics Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression Q Science > QA Mathematics > QA279.5 Bayesian statistical decision theory. |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | MUHAMMAD RAYHAN LAHDJI |
| Date Deposited: | 24 Feb 2022 05:33 |
| Last Modified: | 10 Jun 2024 00:56 |
| URI: | http://repository.its.ac.id/id/eprint/94745 |
Actions (login required)
 |
View Item |