ANALISIS SENTIMEN MEDIA SOSIAL TWITTER MENGENAI KENDARAAN LISTRIK DI INDONESIA MENGGUNAKAN METODE NAÏVE BAYES

BAKAR, M.SYECHAN ABU and Suarli, Pacu Putra (2024) ANALISIS SENTIMEN MEDIA SOSIAL TWITTER MENGENAI KENDARAAN LISTRIK DI INDONESIA MENGGUNAKAN METODE NAÏVE BAYES. Undergraduate thesis, Sriwijaya University.

	Text RAMA_57201_09031281823059.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_57201_09031281823059_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_57201_09031281823059_0018128901_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_57201_09031281823059_0018128901_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (328kB) | Request a copy
	Text RAMA_57201_09031281823059_0018128901_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (103kB) | Request a copy
	Text RAMA_57201_09031281823059_0018128901_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (532kB) | Request a copy
	Text RAMA_57201_09031281823059_0018128901_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (51kB) | Request a copy
	Text RAMA_57201_09031281823059_0018128901_06_ref.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (59kB) | Request a copy
	Text RAMA_57201_09031281823059_0018128901_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy

Abstract

The advancement of technology has led to increasing modernization, consequently raising the demand for fuel with limited fuel resorces. The Indonesian government seeks to address this energy dependency by proposing a transition from conventional vehicles to electric vehicles. The sentimen analysis in this research aims to assess public opinions regarding this proposition. Data collected from Twitter were processed using the Naive Bayes method to generate three sentiment classifications : positive, negative, and neutral. The classification results were then visualized in the form of a wordcloud. From a total of 4515 crawled data, 1236 data remained after the cleaning process. The application of the Naive Bayes method to this data resulted in positive sentiment (900 data), negative (240 data), and neutral sentimen (96 data). Performance testing of the Naive Bayes method was conducted using Cross Validation and the KNN algorithm. With a fold value of 10 in Cross Validation and K value of 7 in KNN, the accuracy value obtained was 79.13%. Precision results were positive = 83.50%, negattive = 61.20%, and neutral = 63.08%. Recall results were positive = 91,67%, negative = 46.67%, and neutral = 42.71%. The results imply that the model has a good level of accuracy with sufficient capability in identifying positive cases and recognizing the majority of true positive cases. These findings indicate a balance between accuracy and the model's ability to handle positive cases.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Classification, Electric Vehicles, Naive Bayes, Sentiment Analysis, Twitter
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 57201-Information Systems (S1)
Depositing User:	M Syechan Abu Bakar
Date Deposited:	25 Jan 2024 06:52
Last Modified:	25 Jan 2024 06:52
URI:	http://repository.unsri.ac.id/id/eprint/139790

Actions (login required)

View Item