twigane-kinyarwanda-teaching-chatbot

Twigane Kinyarwanda Teaching Chatbot

Model Description

This is a Kinyarwanda teaching chatbot model developed as part of the Twigane project. The model is designed to help users learn Kinyarwanda language, culture, and traditions through interactive conversations.

Model Details

• Model Type: Teaching Chatbot
• Language: Kinyarwanda (rw)
• Architecture: Custom trained model with TF-IDF vectorization
• Training Data: Custom Kinyarwanda dataset with 1,010+ samples
• Categories: Vocabulary, Grammar, Culture, Conversation, Translation, Numbers, Family terms

Usage

from transformers import AutoModel
import pickle
import numpy as np

# Load the model components
model = AutoModel.from_pretrained("Maxime-Bakunzi/twigane-kinyarwanda-teaching-chatbot")

# The model includes:
# - Pre-computed embeddings
# - Response texts
# - Metadata
# - TF-IDF preprocessor

Training Data Categories

• Vocabulary: Basic to advanced Kinyarwanda words and meanings
• Grammar: Noun classes, verb conjugations, sentence structure
• Culture: Traditional practices, values, and customs
• Conversation: Greetings, daily interactions, social contexts
• Translation: English-Kinyarwanda translation examples
• Numbers: Counting system and numerical expressions
• Family: Family terms and relationships

Performance

The model provides contextually appropriate responses for Kinyarwanda language learning queries with high accuracy in cultural and linguistic contexts.

Limitations

• Primarily focused on Kinyarwanda language and Rwandan culture
• Best performance on topics covered in training data
• May require additional context for very specific or technical queries

Citation

If you use this model, please cite:

@misc{twigane-teaching-model,
  author = {Twigane Project},
  title = {Kinyarwanda Teaching Chatbot Model},
  year = {2025},
  publisher = {Hugging Face},
  url = {https://huggingface.co/Maxime-Bakunzi/twigane-kinyarwanda-teaching-chatbot}
}