Reimagining service delivery in the agentic era with Google Public Sector
cloud.google.com·19h ago
AI Research
arXiv CS.AI·Jun 9·5 min read
TeamHerald@CHIPSAL 2026: Hate Speech Detection and Sentiment Analysis of Nepali Memes using Transformer-based Architectures and Ensemble Learning
TL;DR
Hate speech detection in Nepali memes is challenging due to code-mixing and limited resources. This study utilized Transformer-based architectures and ensemble learning techniques, specifically Hard and Soft Voting, to analyze hate speech and sentiment.
✦ Why It Matters
Engineers can leverage ensemble learning techniques to improve hate speech detection in multilingual contexts.
Key Takeaways
- Utilize OCR for text extraction from memes.
- Experiment with different Transformer models for classification tasks.
- Apply ensemble methods to improve model performance.
- Focus on task-specific strategies for better results.
How It Works
The study employs Optical Character Recognition (OCR) to extract text from Nepali memes, which are then analyzed using various Transformer-based models. The models are evaluated for their effectiveness in two distinct tasks: detecting hate speech and analyzing sentiment.
The use of ensemble learning, particularly Soft Voting, allows for improved performance by aggregating predictions from multiple models, which is particularly beneficial in multi-class scenarios.