Analisis Sentimen pada X Terkait Kenaikan Harga Bahan Bakar Minyak Menggunakan Metode Machine Learning

Qalbi, Fayha Syifa (2024) Analisis Sentimen pada X Terkait Kenaikan Harga Bahan Bakar Minyak Menggunakan Metode Machine Learning. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 05111940000185-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 July 2026. Download (7MB) | Request a copy

Abstract

X adalah salah satu media sosial yang paling digemari masyarakat sekarang ini. Adanya X mempermudah masyarakat untuk menyebarkan ataupun menerima berita secara cepat dan langsung, sehingga banyak isu terkini yang menjadi hot topic di kalangan masyarakat. Kenaikan harga Bahan Bakar Minyak yang terjadi menjadi salah satu topik yang ramai dibahas di X. Bahan Bakar Minyak sendiri merupakan kebutuhan pokok bagi semua kalangan di masyarakat sehingga perubahan harga terkait BBM akan sangat mempengaruhi banyak faktor di kehidupan sehari - hari. Hadirnya analisis sentimen terkait kasus ini sangat diperlukan untuk memudahkan pemerintah maupun masyarakat menyikapi fenomena yang terjadi. Maka, pada penelitian ini akan dilakukan analisis sentimen terkait kenaikan harga BBM dalam aplikasi X menggunakan metode Machine Learning. Metode machine learning adalah salah satu metode yang banyak digunakan dalam analisis sentimen. Metode digunakan adalah algoritma Naive Bayes Classifier, K-Nearest Neighbor, Support Vector Machine, dan Random Forest. Karena adanya dataset yang tidak seimbang, diterapkan beberapa metode oversampling yaitu SMOTE, ADASYN, SMOTE-Tomek, dan SMOTE-ENN. Setelah dilakukan klasifikasi dengan berbagai model machine learning dan juga metode oversampling dibandingkan seluruh hasil klasifikasi. Hasil klasifikasi tertinggi didapatkan menggunakan model Random Forest dan metode ADASYN pada presentase data training 90% dan data testing 10% dengan akurasi mencapai 91%.
==================================================================================================================================
X is one of the most popular social media in today's society. The existence of X makes it easier for people to spread or receive news quickly and directly, so many current issues have become hot topics among the public. The increase in fuel prices is one of the most discussed topics on X. Fuel itself is a basic need for all people in society, so changes in fuel prices will greatly affect many factors in daily life. The presence of sentiment analysis in this case is needed to make it easier for the government and society to respond to the phenomena that occur. So, in this research, sentiment analysis related to the increase in fuel prices will be carried out in application X using the machine learning method. Machine learning method is one of the methods widely used in sentiment analysis. The methods used are Naive Bayes Classifier algorithm, K-Nearest Neighbor, Support Vector Machine, and Random Forest. Due to the unbalanced dataset, several oversampling methods were applied, namely SMOTE, ADASYN, SMOTE-Tomek, and SMOTE-ENN. After classification with different machine learning models and oversampling methods, all classification results were compared. With a percentage of 90% training data and 10% test data, the Random Forest model and the ADASYN method achieved the best classification results with an accuracy of 91%.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Analisis Sentimen, BBM, K-Nearest Neighbor, Machine Learning, Naive Bayes Classifier; Support Vector Machine, Oversampling, Fuel, Sentiment Analysis
Subjects:	T Technology > T Technology (General)
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Fayha Syifa Qalbi
Date Deposited:	10 Feb 2024 05:41
Last Modified:	10 Feb 2024 05:41
URI:	http://repository.its.ac.id/id/eprint/106666

Actions (login required)

View Item