Masitah, Siti (2024) Penerapan Algoritma Naïve Bayes Untuk Analisis Sentimen Ulasan Film di IMDb:Studi Kasus pada Genre Fantasi, Aksi, dan Horor. Undergraduate thesis, Politeknik Negeri Bengkalis.
![[img]](http://eprints.polbeng.ac.id/style/images/fileicons/text.png) |
Text (Abstract)
SKRIPSI-6304201308-Abstract.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (4MB) |
|
Text (Bab I Pendahuluan)
SKRIPSI-6304201308-Bab I Pendahuluan.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (4MB) |
|
Text (Daftar Pustaka)
SKRIPSI-6304201308-Daftar Pustaka.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (4MB) |
|
Text (Full Text)
SKRIPSI-6304201308-Full Text.pdf - Submitted Version Restricted to Registered users only Available under License Creative Commons Attribution Share Alike. Download (4MB) | Request a copy |
Abstract
This study aims to analyze user sentiment towards three film genres on IMDb: fantasy, action, and horror, using the Naïve Bayes method. It is motivated by the significant role of digital media in transforming the film industry and the emergence of platforms like IMDb as the primary source of information for film enthusiasts. With the increasing ease of internet access and the use of social media, the number of film reviews uploaded online is growing, making film sentiment analysis essential for understanding audience responses. IMDb holds sway in shaping public opinion through user ratings and reviews. Experts have found that IMDb ratings and reviews can influence viewers' perceptions of a film and can be used to predict the success of Hollywood box office films. Therefore, IMDb sentiment analysis provides valuable insights for the film industry to understand audience preferences. This research implements the Naive Bayes Classifier algorithm to classify sentiment in review data. Using the provided dataset, the algorithm achieved an accuracy rate of 0.736, indicating its effectiveness in predicting sentiment. The main advantage of this research lies in the Naive Bayes algorithm's ability to produce satisfactory accuracy, precision, and recall values, making it suitable for real-world applications. However, a limitation exists in the system's performance in identifying information in positive and negative classes, with a success rate of only 78.4%. This is attributed to the imbalance in the number of training data between positive and negative classes, where the negative and neutral classes have fewer data compared to the positive class. Therefore, future research should address the balance of training data to improve the system's performance in sentiment recognition.
| Item Type: | Thesis (Undergraduate) | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||
| Uncontrolled Keywords: | Film reviews, Naive Bayes, IMDB, Jupyter Notebook, Clafication | ||||||||
| Subjects: | 410 ILMU TEKNIK > 450 TEKNIK ELEKTRO DAN INFORMATIKA > 463 Teknik Perangkat Lunak | ||||||||
| Divisions: | Jurusan Teknik Informatika > Sarjana Terapan Rekayasa Perangkat Lunak > TUGAS AKHIR | ||||||||
| Depositing User: | Rekayasa Perangkat Lunak C 2024 | ||||||||
| Date Deposited: | 26 Aug 2024 15:36 | ||||||||
| Last Modified: | 26 Aug 2024 15:36 | ||||||||
| URI: | http://eprints.polbeng.ac.id/id/eprint/13864 |
Actions (login required)
 |
View Item |
