PERANCANGAN FRAMEWORK DETEKSI HOAX PADA MEDIA SOSIAL TWITTER UNTUK SIBER POLRI

• Main Author: Hafif, Muhammad
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/54516
• Institution: Institut Teknologi Bandung

The development of information technology in Indonesia is growing rapidly. The 2018 Indonesian Internet Service Providers Association (APJII) survey explained that Indonesian internet users reached 171.17 million people or 64.8% of the total Indonesian population of 264.16 million people. Internet content (social media) most frequently visited by Indonesians is 50.7% Facebook, 17.8% Instagram, 15.1% Youtube, 1.7% Twitter, and 0.4% Linkedin. The negative impact of using social media is fake news or hoaxes. hoaxes that have a negative impact make people uneasy. The police need to be ready and pro-active in dealing with the threats caused by these hoaxes. Identification of hoax content has been carried out by the internet community who are members of the turnbackhoax.id site. The site is managed by MAFINDO (Indonesian anti hoax society). The method of identification or classification carried out on the turnbackhoax.id site is still done manually, so that if the information is growing, it will be difficult because more information is entered. Previous research on hoaxes was carried out by (Petkovic et al., 2005), (Vukovic et al., 2009), (Chen et al., 2014) and (Rasywir and Purwarianti, 2015), but this research is related to the hoax email domain and classification system experiments for hoax news using the method of Levenshtein Distance, Fuzzy Logic, Feed Forward Neural Network, Naïve Bayes, Support Vector Machine and C4.5 Algorithm. Previous hoax detection still has shortcomings related to not being able to detect and classify tweets that do not include news sites. So that in this study a new measure, measuring instrument and framework can be used by the Police, so that the Cyber Police can act more quickly in preventive efforts.