Irawan, Risnaldy Novendra (2024) Penerapan Principal Component Analysis Pada Analisis Sentimen Menggunakan Multinomial Naive Bayes (Studi Kasus: Pelayanan Publik Kereta Api Lokal DAOP 8). Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
20083010017-cover.pdf Download (4MB) |
|
Text (Bab 1)
20083010017-bab1.pdf Download (615kB) |
|
Text (Bab 2)
20083010017-bab2.pdf Restricted to Repository staff only until 31 May 2026. Download (905kB) |
|
Text (Bab 3)
20083010017-bab3.pdf Restricted to Repository staff only until 31 May 2026. Download (1MB) |
|
Text (Bab 4)
20083010017-bab4.pdf Restricted to Repository staff only until 31 May 2026. Download (2MB) |
|
Text (Bab 5)
20083010017-bab5.pdf Download (518kB) |
|
Text (Daftar pustaka)
20083010017-daftarpustaka.pdf Download (479kB) |
|
Text (Lampiran)
20083010017-lampiran.pdf Restricted to Repository staff only Download (2MB) |
Abstract
This research aims to conduct sentiment analysis on the users of subsidized train services in the operational area 8 Surabaya. The method used for the analysis is the Naïve Bayes Classifier with the objective of understanding the impact of Principal Component Analysis (PCA) feature selection on the Multinomial Naïve Bayes algorithm. The data preprocessing stages included cleaning, weighting, and splitting the data, resulting in a total of 1123 data points with two classes, namely positive and negative. Subsequently, data processing was conducted to find the best PCA features and perform classification. The data processing results using Multinomial Naïve Bayes with manual labeling without PCA feature selection showed more accurate performance compared to using feature selection, achieving an accuracy of 80% during the testing process. The aforementioned scenario also resulted in correctly predicted data of 139 for positive and 40 for negative. Other results on the data showed that the highest accuracy was obtained by the scenario of MNB Classification with PCA 114, reaching 71%, with correctly predicted data of 117 for positive and 42 for negative.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | H Social Sciences > HA Statistics H Social Sciences > HE Transportation and Communications Q Science > Q Science (General) |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Data Science | ||||||||||||
| Depositing User: | Risnaldy Novendra | ||||||||||||
| Date Deposited: | 31 May 2024 02:46 | ||||||||||||
| Last Modified: | 31 May 2024 02:46 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/23656 |
Actions (login required)
 |
View Item |