Summary of Hate Speech Detection and Classification in Amharic Text with Deep Learning, by Samuel Minale Gashe et al.
Hate Speech Detection and Classification in Amharic Text with Deep Learning
by Samuel Minale Gashe, Seid Muhie Yimam, Yaregal Assabie
First submitted to arxiv on: 7 Aug 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel deep learning approach is proposed to detect and classify hate speech in Amharic, a low-resource language. The proposed SBi-LSTM model leverages a custom-annotated dataset of 5k social media posts and comments, categorized into four types: racial, religious, gender-based, and non-hate speech. The model achieves an impressive F1-score performance of 94.8%. This study addresses the gap in hate speech detection for low-resource languages, with potential applications in countries like Ethiopia where hate speech can trigger conflicts. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Hate speech is a big problem on social media, especially in countries with diverse groups. Researchers are working to detect and stop this kind of content. But most of these efforts focus on languages that have lots of data, like English or Spanish. What about languages that don’t have as much data? This is a major gap. To fill it, scientists created a new model called SBi-LSTM that can detect hate speech in Amharic, a language spoken in Ethiopia. They also made a big dataset of 5k social media posts and comments with labels (like “hate” or “not hate”). The model worked really well, getting 94.8% of the answers right! This is an important step forward in stopping hate speech online. |
Keywords
» Artificial intelligence » Deep learning » F1 score » Lstm
