Sistem Pendeteksi Hoax Pada Berita Seputar Pemerintahan Indonesia Berbasis Website Menggunakan Algoritma Naive Bayes

INTAN, ANGGI PANGESTI (2023) Sistem Pendeteksi Hoax Pada Berita Seputar Pemerintahan Indonesia Berbasis Website Menggunakan Algoritma Naive Bayes. Diploma thesis, Institut Teknologi Telkom Purwokerto.

	Text Cover.pdf Download (218kB)
	Text Abstrak.pdf Download (30kB)
	Text Abstrack.pdf Download (30kB)
	Text BAB I.pdf Download (47kB)
	Text BAB II.pdf Download (358kB)
	Text BAB III.pdf Download (156kB)
	Text BAB IV.pdf Restricted to Registered users only Download (407kB) | Request a copy
	Text BAB V.pdf Download (30kB)
	Text Daftar Pustaka.pdf Download (99kB)
	Text Lampiran.pdf Restricted to Registered users only Download (98kB) | Request a copy

Abstract

Hoax news is very widespread in society, this of course can cause bad things such as causing anxiety. In this research a website is offered that is designed to detect hoax news that is spread in the midst of society. This website was designed by utilizing the naïve Bayes algorithm, in which the naïve Bayes algorithm is utilized in estimating the likelihood of several levels of the target variable, with the hope that this research can help the Indonesian people choose which information is true and which is wrong in the Indonesian government. This website can be accessed publicly and freely by all Indonesian people without having to register and pay in advance. The purpose of this study is to find out how the level of accuracy and performance of Naïve Bayes and how the Naïve Bayes algorithm can make it easier for users. This research uses news sourced from kompas.com, detik.com, liputan6.com, and TribunNews.com for true news, and turnbackhoax.id for hoax news that raises issues about government, corruption, parties, the economy, Jokowi, and elections. The research was carried out in several stages, namely the system design stage, manufacture and the last, the system testing stage. The tools and materials used in this study include the python programming language, visual studio code, datasets collected from news websites and also the turnbackhoax.id website, the Google colaboratory which is used to store program code. The turnbackhoax.id website is a site managed by Mafindo, the Indonesian Anti Hoax Society. Classification of news validity using the naïve Bayes method produces good accuracy on data from the website and produces accuracy values with the highest percentage of 91%, 11% precision, and 24% recall, this means that it shows very good naïve Bayes performance. Keywords: hoax news, Indonesian government, Naïve Bayes detector, Natural Language Processing

Actions (login required)

View Item