🚀 IFRI FAQ Chatbot - A Smart Revolution for Education! 🎓💡
(Github link to the full project: https://github.com/FredyHoundayi/IFRI-FAQ-Chatbot/tree/master )
This project represents a major breakthrough in accessing academic information through artificial intelligence. By combining Chainlit , Hugging Face , and an LLM model (LLaMA 2) optimized with a RAG search engine, it pushes the boundaries of digital assistance.
✨ Innovation in the service of knowledge : This chatbot goes far beyond traditional FAQs. It interprets, analyzes and synthesizes relevant information from PDF documents, thus providing precise and contextual answers to IFRI-UAC students and researchers.
🌍 A significant impact : With this technology, access to academic resources becomes instantaneous and intuitive , reducing the barriers associated with searching for information. This is a giant step towards democratizing knowledge and optimizing the university experience.
This project is a shining example of the power of AI applied to education , opening the way to new perspectives for teaching and research! 🚀📖
✨ Features
✅ Enhanced Search (RAG) : Scans PDF files for accurate answers.
✅ Advanced Model (LLaMA 2 - 7B HF) : Intelligent text generation.
✅ Conversational Memory : Context tracking for fluid exchanges.
✅ Dark and light mode : Adaptive logos.
✅ Easy deployment : Compatible with Hugging Face Spaces .
🚀 Local Installation and Execution
🔹 1. Clone the project
git clone https://huggingface.co/spaces/ton-utilisateur/ton-projet cd ton-projet
🔹 2. Install dependencies
pip install -r requirements.txt
🔹 3. Add PDF files
Place your documents in the data/ folder.
🔹 4. Launch the application
chainlit run app.py
The interface will be accessible at http://localhost:8000 🌍.
🏗️ Project Structure
ifri_faqchatbot/
│── app.py # Code principal du chatbot
│── requirements.txt # Dépendances à installer
│── data/ # 📂 Dossier contenant les fichiers PDF
│ ├── Information IFRI.pdf
│
│── public/ # 📂 Dossier pour les logos et favicon
│ ├── logo_light.png
│ ├── logo_dark.png
│ ├── favicon.png
│──demo_video.zip
│── README.md # Documentation du projet
🔧 Personalization
🔹 Change the LLM model :
Modify the following line in app.py to test for another pattern:
llm = HuggingFaceEndpoint(repo_id="meta-llama/Llama-2-7b-hf", task="text-generation")
You can try Mistral , Falcon , or GPT-J .
🔹 Edit logos and favicon :
Replaces logo_light.png, logo_dark.png, and favicon.png in the public/ folder.
🤝 Contribute
Contributions are welcome!
If you want to add features , make a pull request or open an issue .
📜 License
This project is licensed under the MIT License.
You can use and modify it freely. 🚀
🎉 Thank you for using IFRI FAQ Chatbot!
If you have any questions, don't hesitate to ask me . 😃
