Application of LSTM and GloVe Word Embedding for Hate Speech Detection in Indonesian Twitter Data

Imaduddin, Helmi; Kusumaningtias, Lucky Anggari; A'la, Fiddin Yusfida

doi:10.18280/isi.280430

Cited by 2 publications

(2 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning models: In the realm of deep learning methodologies, we incorporate two variations of RNN, specifically, the LSTM model [37] and the GRU model [38]. The application of this model in Bahasa Indonesea has been recognized beneficial [39], [40]. The architectural blueprint consists of multiple tiers, initiating with a 300-dimensional embedding layer.…”

Section: Experimental Settingsmentioning

confidence: 99%

Enhancing hate speech detection in Indonesian using abusive words lexicon

Wahyu Pamungkas,

Purworini,

Galih Prasetyo Putri

et al. 2024

IJEECS

View full text Add to dashboard Cite

Hate speech is a major challenge in Indonesia, a diverse country with multiple languages and a dynamic online landscape. This research explores the phenomenon of hate speech and its detection, particularly in language contexts with limited resources. We introduce a new abusive words lexicon, created by collecting words from various sources, adapted for Indonesian, Javanese and Sundanese. Our study investigates the practical implementation of this lexicon. We conducted extensive experiments using different datasets and machine learning models, aiming to improve hate speech detection. The results consistently show a positive impact of the lexicon, which significantly improves detection, especially in languages with fewer resources. But this research paves the way for further exploration. The lexicon can be expanded, broadening its scope. Additionally, we suggest investigating more sophisticated models, such as transformerbased models, to more effectively detect hate speech. In a world where hate speech is a growing problem, our research provides valuable insights and tools to combat it effectively in Indonesia and other countries.

show abstract

Section: Experimental Settingsmentioning

confidence: 99%

Enhancing hate speech detection in Indonesian using abusive words lexicon

Wahyu Pamungkas,

Purworini,

Galih Prasetyo Putri

et al. 2024

IJEECS

View full text Add to dashboard Cite

show abstract

“…Various types of neural networks, including CNN, RNN, and LSTM [24], have been utilized with considerable success across numerous fields [20,25]. In the proposed system, the LSTM model, which forms connections with preceding states in a sequence and addresses the vanishing gradient problem using specialized memory cells, has been incorporated to improve attack detection in SDN [25].…”

Section: Deep Learningmentioning

confidence: 99%

Enhanced Intrusion Detection in Software-Defined Networks Through Federated Learning and Deep Learning

Abd Al-Ameer,

Bhaya

2023

ISI

View full text Add to dashboard Cite

Software-defined networks (SDNs), while offering a revolutionary global view of the network, remain susceptible to a variety of attacks. This vulnerability necessitates innovative solutions for preserving data privacy and enhancing network security. The work presented herein introduces an innovative network anomaly detection methodology leveraging both federated learning (FL) and deep learning (DL) techniques. In contrast to traditional collaborative learning, where potential privacy compromises arise from the distribution of local training data to a central server, the proposed methodology enables each switch in the network to collect data from its connected hosts and independently train a local Long Short-Term Memory (LSTM) model. Subsequently, each switch encrypts and forwards its model parameters to the controller. Upon receipt, the controller decrypts the parameters from each switch, computes their average, and formulates a global LSTM model. This model is disseminated to every switch in the network, enabling each host to retrain its local model according to the global parameters. This iterative process is conducted multiple times to maintain the timeliness of the information. Evaluation of the proposed methodology using the UNSW-NB15 dataset, in conjunction with NF-UQ-NIDS-v2 and CICIDS2017 datasets, demonstrated its efficacy in anomaly detection, with performance exceeding a 96.75% accuracy rate.

show abstract

Application of LSTM and GloVe Word Embedding for Hate Speech Detection in Indonesian Twitter Data

Cited by 2 publications

References 20 publications

Enhancing hate speech detection in Indonesian using abusive words lexicon

Enhancing hate speech detection in Indonesian using abusive words lexicon

Enhanced Intrusion Detection in Software-Defined Networks Through Federated Learning and Deep Learning

Contact Info

Product

Resources

About