Hate Speech Detection on Social Media using Bidirectional Long Short Term Memory (BiLSTM)

ISMI NUR AZIZA

Informasi Dasar

110 kali
23.05.067
006.31
Karya Ilmiah - Thesis (S2) - Reference

The uses of Social Media give some benefits for human to interaction in society. But, sometimes the information dissemination carried out by an individual or group cannot be justified and causes abuse on social media. This Research is about detecting hatespeech on social media, especially Instagram comment section and also Twitter using variant of Reccurent Neural Network, that is Bidirectional LSTM. Recurrent Neural Network is a type of neural network that contains loops, allowing information to be stored within the network. This objective of this research is to identifying hate speech on social media, especially instagram and twitter platform with several aspect such as Abusive, Individual, and Other. The result of classifying Hate Speech using LSTM and fasttext achieve 81% for f1 score and on aspect based hate speech, Abusive aspect achieve higher result 84% on f1 score.

Subjek

Machine Learning
 

Katalog

Hate Speech Detection on Social Media using Bidirectional Long Short Term Memory (BiLSTM)
 
 
Indonesia

Sirkulasi

Rp. 0
Rp. 0
Tidak

Pengarang

ISMI NUR AZIZA
Perorangan
Arie Ardiyanti Suryani
 

Penerbit

Universitas Telkom, S2 Informatika
Bandung
2023

Koleksi

Kompetensi

 

Download / Flippingbook

 

Ulasan

Belum ada ulasan yang diberikan
anda harus sign-in untuk memberikan ulasan ke katalog ini