HASAN, DHAIFULLAH FADHIL (2025) FILMFEELS: ANALISIS SENTIMEN REVIEW FILM MENGGUNAKAN ALGORITMA NAÏVE BAYES,DECISION TREE, LOGISTIC REGRESSION, DAN ENSEMBLE LEARNING. S1 thesis, Universitas Mercu Buana Jakarta - Menteng.
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (COVER)
41520120040-Dhaifullah Fadhil Hasan-01 Cover - DHAI FULLAH FADHIL HASAN.pdf Download (434kB) |
|
Text (BAB I)
41520120040-Dhaifullah Fadhil Hasan-Bab 1 - DHAI FULLAH FADHIL HASAN.pdf Restricted to Registered users only Download (58kB) |
|
Text (BAB II)
41520120040-Dhaifullah Fadhil Hasan-Bab 2 - DHAI FULLAH FADHIL HASAN.pdf Restricted to Registered users only Download (213kB) |
|
Text (BAB III)
41520120040-Dhaifullah Fadhil Hasan-Bab 3 - DHAI FULLAH FADHIL HASAN.pdf Restricted to Registered users only Download (193kB) |
|
Text (BAB IV)
41520120040-Dhaifullah Fadhil Hasan-Bab 4 - DHAI FULLAH FADHIL HASAN.pdf Restricted to Registered users only Download (809kB) |
|
Text (BAB V)
41520120040-Dhaifullah Fadhil Hasan-Bab 5 - DHAI FULLAH FADHIL HASAN.pdf Restricted to Registered users only Download (37kB) |
|
Text (DAFTAR PUSTAKA)
41520120040-Dhaifullah Fadhil Hasan-Daftar Pustaka - DHAI FULLAH FADHIL HASAN.pdf Restricted to Registered users only Download (125kB) |
|
Text (LAMPIRAN)
41520120040-Dhaifullah Fadhil Hasan-Lampiran - DHAI FULLAH FADHIL HASAN.pdf Restricted to Registered users only Download (708kB) |
Abstract
Penelitian ini mengembangkan sistem analisis sentimen review film menggunakan Ensemble Learning dengan algoritma Naïve Bayes, Decision Tree, dan Logistic Regression. Metode pengembangan sistem menggunakan pendekatan incremental dengan dua tahap utama, yaitu preprocessing untuk persiapan data dan klasifikasi sentimen pada 397.146 data review film. Hasil evaluasi menunjukkan bahwa model AdaBoost dengan Decision Tree sebagai estimator dasar memperoleh akurasinya sebesar 76.8%, dengan performa sangat baik pada kelas Positif (recall 96% dan F1-score 0.86). Namun, model ini kesulitan dalam mengidentifikasi kelas Negatif, dengan precision 0.67 dan recall 0.22. Selanjutnya, model Voting Classifier yang menggabungkan DecisionTreeClassifier, LogisticRegression, dan MultinomialNB dengan pendekatan soft voting dan pembobotan menghasilkan akurasinya 82.23%, dengan performa terbaik pada kelas Positif (presisi 0.83, recall 0.96, F1-score 0.89), namun kinerjanya kurang optimal pada kelas Negatif (presisi 0.79, recall 0.44, F1-score 0.56). This study develops a sentiment analysis system for movie reviews using Ensemble Learning with Naïve Bayes, Decision Tree, and Logistic Regression algorithms. The system development method employs an incremental approach with two main stages: preprocessing for data preparation and sentiment classification of movie reviews. Evaluation results show that the AdaBoost model with Decision Tree as the base estimator achieved an accuracy of 76.8%, with excellent performance on the Positive class (recall 96% and F1-score 0.86). However, the model struggled to identify the Negative class, with precision 0.67 and recall 0.22. Furthermore, the Voting Classifier model, combining DecisionTreeClassifier, LogisticRegression, and MultinomialNB with a soft voting approach and weighting, achieved an accuracy of 82.23%, with the best performance on the Positive class (precision 0.83, recall 0.96, F1-score 0.89). However, its performance on the Negative class was less optimal (precision 0.79, recall 0.44, F1-score 0.56). The system was tested on a movie review dataset consisting of 397,146 entries and successfully provided ease of use, prediction accuracy, fast access, and improved information visualization. The parameters analyzed include accuracy, precision, recall, and F1-score for each algorithm.
| Item Type: | Thesis (S1) |
|---|---|
| NIM/NIDN Creators: | 41520120040 |
| Uncontrolled Keywords: | Sentimen Review Film, Ensemble Learning, Naïve Bayes, Decision Tree, Logistic Regression Sentimen Review Film, Ensemble Learning, Naïve Bayes, Decision Tree, Logistic Regression |
| Subjects: | 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 004 Data Processing, Computer Science/Pemrosesan Data, Ilmu Komputer, Teknik Informatika |
| Divisions: | Fakultas Ilmu Komputer > Informatika |
| Depositing User: | NAIMAH NUR ISLAMIDIYANAH |
| Date Deposited: | 23 Jun 2025 03:43 |
| Last Modified: | 23 Jun 2025 03:43 |
| URI: | http://repository.mercubuana.ac.id/id/eprint/95866 |
Actions (login required)
 |
View Item |
