Detecting Hate Speech In Twitter Using Long Short-Term Memory and Naïve Bayes Method

Firman Sriyono (AMIKOM University Yogyakarta, Indonesia)
Kusrini Kusrini (AMIKOM University Yogyakarta, Indonesia)
Asro Nasiri (AMIKOM University Yogyakarta, Indonesia)

Syntax Literate: Jurnal Ilmiah Indonesia
Jurnal Ilmiah Indonesia

The information technologi’s development has been very sophisticated and easy, so that it becomes a lifestyle for people throughout the world without exception Indonesia which also affected by the development of this technology. One of the benefits of information technology is the emergence various kinds of social networking sites or social media such as Facebook, Twitter and Instagram. Technological developments isn’t only have a positive impact, but also have a negative impact the crime of insult or hate speech. This study is aims to classify Indonesian hate speech sentences based on hate speech and neutral sentiments using the Long Short-Term Memory (LSTM) method. Research data is obtained from Indonesian-language tweets. In testing process, the LSTM method will be compared with the Naïve Bayes method