EUVIMEDEUVIMED entdecken
Crossref

Sustainable AI solutions: A review of green technology for hate speech detection in vídeos

Ryan Ari Setyawan, Herman Dwi Surjono, Ratna Wardani

Multidisciplinary Reviews · 10.07.2026 · 10 (1) · 2027014

DOI: 10.31893/multirev.2027014 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Multidisciplinary Reviews, online erschienen am 10.07.2026. Verfasst von Ryan Ari Setyawan und 2 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Hate speech, which targets individuals or groups based on characteristics such as race, ethnicity, gender, or religion, remains a pervasive problem on social media platforms. The absence of effective filtering systems highlights the need for advanced AI-driven solutions to automatically detect hate speech in video content. This study conducts a literature review of AI algorithms suitable for this purpose, emphasizing the importance of minimizing energy consumption in line with green technology principles. The review identifies three key algorithms: Convolutional Neural Networks (CNN), Bidirectional Encoder Representations from Transformers (BERT), and Long Short-Term Memory networks (LSTM). CNNs are particularly adept at analyzing visual content and detecting harmful images or videos associated with hate speech. In contrast, BERT and LSTM, both powerful natural language processing models, excel at understanding spoken or written language within videos. By combining these models into a multimodal framework—processing both textual and visual inputs—detection accuracy can be significantly improved. A key finding of the study is the need to align AI model development with green technology principles, focusing on reducing the energy footprint of computational processes. Optimizing these algorithms to minimize CPU usage and employing efficient architectures can significantly reduce their environmental impact. Techniques such as model pruning, quantization, or fusion methods can further optimize performance while lowering energy consumption. This approach ensures that AI-based hate speech detection systems are not only more accurate but also more sustainable. By developing energy-efficient models, the study proposes a solution that balances effectiveness in content moderation with a reduced environmental impact, paving the way for more eco-friendly and efficient AI applications.

Zitieren

APA

Setyawan, R. A., Surjono, H. D., & Wardani, R. (2026). Sustainable AI solutions: A review of green technology for hate speech detection in vídeos. Multidisciplinary Reviews, 10(1), 2027014. https://doi.org/10.31893/multirev.2027014

Vancouver

Setyawan RA, Surjono HD, Wardani R. Sustainable AI solutions: A review of green technology for hate speech detection in vídeos. Multidisciplinary Reviews. 2026;10(1):2027014. doi:10.31893/multirev.2027014

BibTeX
@article{Setyawan2026,
  author = {Setyawan, Ryan Ari and Surjono, Herman Dwi and Wardani, Ratna},
  title = {Sustainable AI solutions: A review of green technology for hate speech detection in vídeos},
  journal = {Multidisciplinary Reviews},
  year = {2026},
  volume = {10},
  number = {1},
  pages = {2027014},
  doi = {10.31893/multirev.2027014}
}
Kostenlos starten →