SISTEM DETEKSI BERITA HOAX PEMILU 2024 INDONESIA MENGGUNAKAN ALGORITMA KNN DAN SVM

DickiPrabowo, Reza (2024) SISTEM DETEKSI BERITA HOAX PEMILU 2024 INDONESIA MENGGUNAKAN ALGORITMA KNN DAN SVM. Skripsi (S1) thesis, Universitas Muhammadiyah Ponorogo.

	Text (Surat Persetujuan Unggah Karya Ilmiah) UNIVERSITAS MUHAMMADIYAH PONOROGO.pdf Download (290kB)
	Text (HALAMAN DEPAN) hlmn depan.pdf Download (3MB)
	Text (BAB l) bab 1.pdf Download (1MB)
	Text (BAB ll) bab 2.pdf Restricted to Registered users only Download (1MB)
	Text (BAB lll) bab 3.pdf Restricted to Registered users only Download (1MB)
	Text (BAB IV) bab 4.pdf Restricted to Registered users only Download (2MB)
	Text (BAB V) bab 5.pdf Restricted to Registered users only Download (1MB)
	Text (DAFTAR PUSTAKA) daftar pstaka.pdf Download (1MB)
	Text (SKRIPSI FULL TEXT) skripsi.pdf Restricted to Registered users only Download (4MB)

Abstract

Technological advances in the current era have had a significant impact, both positive and negative. The phenomenon of false
information or hoaxes is increasingly rampant, especially in the political context. The spread of hoax information has become
a strategy commonly used in political contestations in various countries, including in general elections. By looking at the
increasing trend of political hoax attacks before, during and after the election, this research highlights the urgency of dev eloping
more sophisticated and accurate hoax detection technology. Previous research has been carried out using identification methods
but it is considered not optimal, because the resulting accuracy is still not high. This research aims to use a combined meth od
of the KNN and SVM algorithms to develop a more effective hoax news detection system using the two combined methods.
The combined method that has been carried out shows very good accuracy, reaching around 93.31%. KNN provides advantages
in its simplicity and ability to capture non-linear patterns, while SVM excels in performance in high-dimensional spaces by a
large margin. By combining the strengths of different algorithms, model stacking produces a more accurate and effective model
than using one algorithm.

Item Type:	Thesis (Skripsi (S1))
Subjects:	H Social Sciences > H Social Sciences (General) T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:	Faculty of Engineering
Depositing User:	ft . userft
Date Deposited:	03 Sep 2024 07:52
Last Modified:	06 Jan 2025 02:32
URI:	https://eprints.umpo.ac.id/id/eprint/14848

Actions (login required)

View Item