Artificial Intelligence Makes a Leap in the Fight Against Online Hate Speech in Portuguese
A new study reveals that artificial intelligence is becoming increasingly adept at identifying hate speech on social media in the Portuguese language.
Computer researchers have developed state-of-the-art language models capable of automatically detecting offensive or discriminatory content on platforms like YouTube and Twitter.
Although previous research had already explored this topic, research in Portuguese was still scarce. However, the new study, which used models such as BERTimbau and GPT-3.5, obtained promising results, with accuracy rates exceeding 87% in some cases.
What does this mean?
- More efficient moderation: Online platforms will be able to moderate their content more effectively, quickly removing hate speech.
- Safer communities: Users will be able to have a safer and more respectful online experience.
- Continuous advancements: Research in this area continues to advance, and we can expect hate speech detection tools to become even more accurate in the future.
Researchers emphasize the importance of continuing to invest in research in this area, especially in underrepresented languages like Portuguese. The automatic detection of hate speech is a crucial step in creating a healthier and more inclusive online environment.
For more information, see the full article: https://ieeexplore.ieee.org/document/10602515