Ivtytah Ein, . (2022) ANALISIS SENTIMEN TERHADAP LAYANAN TRANSJAKARTA PADA MEDIA SOSIAL INSTAGRAM MENGGUNAKAN METODE NAÏVE BAYES DAN SELEKSI FITUR INFORMATION GAIN. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (50kB) |
|
Text
AWAL.pdf Download (844kB) |
|
Text
BAB 1.pdf Download (138kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (316kB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (190kB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (797kB) |
|
Text
BAB 5.pdf Download (60kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (188kB) |
|
Text
DAFTAR RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (63kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (16MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (501kB) |
Abstract
PT. Jakarta Transportation (Transjakarta) responds to the DKI Jakarta Provincial Government's policy regarding restrictions on transportation modes by limiting bus fleets, travel routes, operating hours and transport capacity. On October 22, 2021, Transjakarta announced that services would return to normal. Even though the service is operating normally again, there are still many transjakarta users who express criticism and opinions related to transjakarta services on their instagram account. This study aims to build a sentiment classification model using the naïve bayes method and selection of information gain features on public opinion regarding transjakarta services on Instagram social media. Comment data is divided into positive class and negative class based on manual labeling carried out by 3 annotators which later on the data will be pre-processed, TF-IDF weighting, feature selection, and data sharing of 70% training data using the SMOTE method and 30% test data before entering the modeling stage. In this study, there are two models, namely the nave Bayes classification model without feature selection and with feature selection. The evaluation results for the nave Bayes classification model obtained an accuracy of 81.42%, recall of 69.64%, precision of 63.93%, and specificity of 85.71%. While the results of the nave Bayes classification model with information gain obtained an accuracy of 86.66%, recall of 71.42%, precision of 76.92%, and specificity of 92.20%.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No Panggil : 1810511010] [Pembimbing 1 : Iin Ernawati] [Pembimbing 2 : Yuni Widiastiwi] [Penguji 1 : Jayanta] [Penguji 2 : Nurul Chamidah] |
| Uncontrolled Keywords: | Sentiment, Instagram, Naïve Bayes, Feature Selection, Information Gain |
| Subjects: | T Technology > T Technology (General) |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
| Depositing User: | Ivtytah Ein |
| Date Deposited: | 11 Aug 2022 06:26 |
| Last Modified: | 11 Aug 2022 06:26 |
| URI: | http://repository.upnvj.ac.id/id/eprint/19716 |
Actions (login required)
 |
View Item |