SEPTIA, PRANA (2021) ANALISIS SENTIMEN PENGGUNA YOUTUBE DI INDONESIA PADA CHANNEL OTOMOTIF NAÏVE BAYES CLASSIFIER. Tugas Akhir thesis, University of Technology Yogyakarta.
|
Text
5170411184_PRANA SEPTIA_ABSTRAK.pdf Download (10kB) | Preview |
Abstract
ANALYSIS OF SENTIMENT OF YOUTUBE USERS IN INDONESIA ON NAÏVE BAYES CLASSIFIER AUTOMOTIVE CHANNEL ABSTRACT Currently, social media has become a very popular communication tool in Indonesia. One of these social media is Youtube. Youtube is a platform to provide education, reviews, information, and entertainment in the form of videos, one of which is the presentation of educational videos in the automotive sector. This study aims to determine the level of sentiment towards the comments of Youtube site users on automotive channels quickly and precisely. In this study, sentiment analysis is a process of classifying textual documents into two classes, namely the negative and positive sentiment classes. Furthermore, the data is preprocessed which consists of casefolding, removing numbers and characters, tokenizing, stopword removal and stemming data. After preprocessing, the TF-IDF is weighted. To find out the classification of each sentiment in the comments, the Naïve Bayes Classifier method is used, namely by calculating the conditional probability of each word weight from the comment and calculating the prior probability to determine the positive or negative sentiment analysis results from the comments given. From the research that has been done, researchers have succeeded in making a sentiment analysis system with an average accuracy of 90.5% from 800 training data and 200 test data. Keywords: Sentiment Analysis, Naïve Bayes Classifier, Youtube.
| Item Type: | Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir) |
|---|---|
| Subjects: | T Technology > T Technology (General) |
| Divisions: | Fakultas Sains Dan Teknologi > S1 Informatika |
| Depositing User: | Kaprodi S1 Informatika UTY |
| Date Deposited: | 28 Oct 2021 03:47 |
| Last Modified: | 28 Oct 2021 03:47 |
| URI: | http://eprints.uty.ac.id/id/eprint/8576 |
Actions (login required)
 |
View Item |
