PERBANDINGAN PERFORMA ALGORITMA SUPPORT VECTOR MACHINE (SVM) DAN CONVOLUTIONAL NEURAL NETWORK (CNN) PADA ANALISIS SENTIMEN KONFLIK PALESTINA ISRAEL DI X

Maulana Wirayudha, - (2024) PERBANDINGAN PERFORMA ALGORITMA SUPPORT VECTOR MACHINE (SVM) DAN CONVOLUTIONAL NEURAL NETWORK (CNN) PADA ANALISIS SENTIMEN KONFLIK PALESTINA ISRAEL DI X. S1 thesis, Universitas Pendidikan Indonesia.

	Text S_PSTI_2004991_Title.pdf Download (494kB)
	Text S_PSTI_2004991_Chapter1.pdf Download (114kB)
	Text S_PSTI_2004991_Chapter2.pdf Restricted to Staf Perpustakaan Download (544kB)
	Text S_PSTI_2004991_Chapter3.pdf Download (297kB)
	Text S_PSTI_2004991_Chapter4.pdf Restricted to Staf Perpustakaan Download (722kB)
	Text S_PSTI_2004991_Chapter5.pdf Download (90kB)
	Text S_PSTI_2004991_Appendix.pdf Restricted to Staf Perpustakaan Download (1MB)

Abstract

Konflik Palestina-Israel yang berlangsung sejak 1948 terus menarik perhatian internasional, termasuk Indonesia. Konflik ini dipicu oleh perebutan wilayah, pembentukan negara Israel, dan serangan dari kedua belah pihak, serta menjadi topik hangat di media sosial X (Twitter) dengan jutaan postingan. Perdebatan ini menciptakan pandangan sosial beragam di Indonesia, baik positif, negatif, maupun netral. Penelitian ini bertujuan memahami persepsi dan sentimen masyarakat Indonesia terhadap konflik Palestina-Israel melalui analisis sentimen di media sosial X menggunakan model machine learning SVM dan CNN. Penelitian dilakukan melalui tahapan crawling data, text preprocessing, labelling, modelling, hingga evaluasi performa model dengan berbagai alat ukur. Hasil distribusi sentimen dari 1225 tweet menunjukkan 66% sentimen negatif, 25% positif, dan 9% netral. Evaluasi performa model SVM menunjukkan akurasi 80%, sementara model CNN menunjukkan akurasi 77%. Penelitian ini menemukan reaksi beragam dari pengguna X terkait konflik ini yang cenderung negatif, serta model SVM menghasilkan akurasi lebih baik dibandingkan CNN, meskipun kurang efektif dalam menganalisis sentimen netral. ----- The Palestine-Israel conflict, which has been ongoing since 1948, continues to attract international attention, including from Indonesia. This conflict is triggered by territorial disputes, the establishment of the state of Israel, and attacks from both sides, and it has become a hot topic on social media X (Twitter) with millions of posts. This debate creates diverse social viewpoints in Indonesia, including positive, negative, and neutral perspectives. This research aims to understand Indonesian public perception and sentiment towards the Palestine-Israel conflict through sentiment analysis on social media X using SVM and CNN machine learning models. The research was conducted through the stages of data crawling, text preprocessing, labeling, modeling, and performance evaluation using various measurement tools. The sentiment distribution results from 1225 tweets show 66% negative sentiment, 25% positive, and 9% neutral. The performance evaluation of the SVM model shows an accuracy of 80%, while the CNN model shows an accuracy of 77%. This study finds diverse reactions from X users regarding this conflict, which tend to be negative, and the SVM model demonstrates better accuracy compared to CNN, although it is less effective in analyzing neutral sentiment.

Item Type:	Thesis (S1)
Additional Information:	https://scholar.google.com/citations?user=9kIPRscAAAAJ&hl=en&authuser=2 ID Sinta Dosen Pembimbing: Suprih Widodo: 5978120 Liptia Venica: 6779029
Uncontrolled Keywords:	Palestina, Israel, Analisis Sentimen, SVM, CNN Palestine, Israel, Sentiment Analysis, SVM, CNN
Subjects:	L Education > L Education (General) T Technology > T Technology (General)
Divisions:	UPI Kampus Purwakarta > S1 Pendidikan Sistem Teknologi dan Informasi
Depositing User:	Maulana Wirayudha -
Date Deposited:	29 Jul 2024 06:34
Last Modified:	29 Jul 2024 09:22
URI:	http://repository.upi.edu/id/eprint/119339

Actions (login required)

View Item