Komparasi Akurasi Ekstraksi Fitur Metode GloVe dan FastText Menggunakan Convolutional Neural Network pada Analisis Sentimen Mengenai Pemindahan Ibu Kota Negara Indonesia

Putri, Desya Ristya (2024) Komparasi Akurasi Ekstraksi Fitur Metode GloVe dan FastText Menggunakan Convolutional Neural Network pada Analisis Sentimen Mengenai Pemindahan Ibu Kota Negara Indonesia. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 20081010240.-cover.pdf Download (1MB)
	Text (Bab 1) 20081010240.-bab1.pdf Download (152kB)
	Text (Bab 2) 20081010240.-bab2.pdf Restricted to Repository staff only until 12 July 2026. Download (326kB)
	Text (Bab 3) 20081010240.-bab3.pdf Restricted to Repository staff only until 12 July 2026. Download (830kB)
	Text (Bab 4) 20081010240.-bab4.pdf Restricted to Repository staff only until 12 July 2026. Download (1MB)
	Text (Bab 5) 20081010240.-bab5.pdf Download (141kB)
	Text (Daftar Pustaka) 20081010240.-daftarpustaka.pdf Download (147kB)

Abstract

Currently, social media has become a popular virtual communication platform among the public. It contains a variety of information that often includes opinions. However, this frequently leads to an imbalance between the information conveyed and how it is interpreted by readers. Not infrequently, the information spread also includes hoaxes, complicating the understanding of the actual sentiment intended to be conveyed. Especially in recent years, accurate information on governmental topics has become quite a discussion point. This research aims to assess the accuracy of machine learning experiments by applying GloVe and FastText word embedding methods in analyzing sentiment towards YouTube user comments regarding the relocation of Indonesia's capital city. These methods were tested by applying a Convolutional Neural Network (CNN) algorithm on 44,957 comment data, divided into training and testing data with a ratio of 70:30. The experimental results showed that using GloVe-CNN yielded an accuracy of 76.1%, while FastText-CNN yielded an accuracy of 75.5%. Thus, this research highlights the importance of feature extraction and sentiment analysis using methods such as GloVe and FastText in classification.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Puspaningrum, Eva Yulia	NIDN0005078908	evapuspaningrum.if@upnjatim.ac.id
Thesis advisor	Maulana, Hendra	NIDN1423128301	hendra.maulana.if@upnjatim.ac.id

Subjects:

Q Science > Q Science (General)
T Technology > T Technology (General) > T385 Computer Graphics

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Desya Ristya Putri

Date Deposited:

12 Jul 2024 09:37

Last Modified:

12 Jul 2024 09:37

URI:

https://repository.upnjatim.ac.id/id/eprint/25931