SamuelAdedeji190/Yoruba-Emoji-Translation-System

Bi-directional Yoruba ↔ Emoji machine translation system built with Python and Streamlit. First of its kind, developed as a postgraduate project using hybrid NLP methods. # Yoruba-Emoji-Translation-System Bi-directional Yoruba ↔ Emoji machine translation system built with Python and Streamlit. First of its kind, developed as a postgraduate project using hybrid NLP methods. Yoruba ↔ Emoji Translator A bi-directional Yoruba ↔ Emoji machine translation system built with Python and Streamlit. This is the first of its kind, developed as a postgraduate project using hybrid NLP techniques. ## Features: • Translate Yoruba text into Emojis and vice versa. • Bi-directional translation: Yoruba ↔ Emoji. • Built with Python and an interactive Streamlit app. • Implements hybrid NLP methods for accurate translations. • User-friendly interface suitable for research and demonstration. ## Tech Stack - Python (pandas, scikit-learn) - Streamlit (web interface) - CSV Dataset for scalability and transparency ## Dataset - Yoruba → Emoji mappings (`yoruba_emoji_100.csv`) - Emoji → Yoruba mappings (`emoji_to_yoruba_100.csv`) ## Example entries: | Yoruba | Emoji | |-------------|---------| | inú mi dùn | 😀 | | ọmọ | 👶 | | ọjọ́ | 📅 | ## Accuracy - Yoruba → Emoji: **95%** - Emoji → Yoruba: **92%** - Evaluated on a curated 100-word dataset + human feedback. ## Demo 👉 Streamlit App Link: yoruba-emoji-translation-sy… ## Run locally with: bash streamlit run app.py