ALLAM-RAG: A Context-Aware Saudi Arabic Conversational AI System for Alzheimer’s Support
Overview
ALLAM-RAG is a Saudi Arabic conversational AI system that combines Retrieval-Augmented Generation (RAG) with the ALLAM Large Language Model to provide context-aware assistance for Alzheimer's patients and caregivers.
The system retrieves relevant information from specialized knowledge bases using hybrid vector search and reranking before generating accurate, personalized responses through the ALLAM language model. By combining retrieval with language generation, the system minimizes hallucinations while delivering reliable and natural Arabic conversations tailored to the Saudi dialect.
Features
- Saudi Arabic conversational assistant
- Retrieval-Augmented Generation (RAG) architecture
- Powered by the ALLAM Large Language Model
- Hybrid semantic and keyword retrieval using Weaviate
- Multilingual reranking using Cohere Rerank
- Context-aware response generation
- Alzheimer's patient support
- General Saudi knowledge support
- Real-time time and day awareness
- Arabic text normalization and intent detection
System Architecture
User Query
│
▼
Query Classification
│
▼
Knowledge Base Selection
│
▼
Weaviate Hybrid Search
│
▼
Cohere Reranker
│
▼
Relevant Context
│
▼
ALLAM Language Model
│
▼
Generated Response
Technologies Used
- Python
- ALLAM-7B-Instruct
- Transformers
- Hugging Face
- Weaviate Vector Database
- Cohere Rerank API
- NumPy
Knowledge Bases
The system utilizes two specialized knowledge bases:
Alzheimer's Knowledge Base
Contains information related to:
- Patient identity
- Memory assistance
- Medication reminders
- Daily activities
- Alzheimer's symptoms
- Caregiving guidance
- Orientation support
General Knowledge Base
Contains Saudi Arabic information including:
- Islamic knowledge
- Saudi Arabia
- Culture
- Geography
- Public information
Retrieval Pipeline
The retrieval pipeline consists of:
- Query preprocessing and normalization
- Query classification
- Knowledge base selection
- Hybrid retrieval from Weaviate
- Cohere multilingual reranking
- Context selection
- Response generation using ALLAM
Example Queries
Alzheimer's Support
- مين أنا؟
- وين ساكن؟
- كيف حالتي الصحية؟
- نسيت أخذ الدواء
- ليه أنا هنا؟
Project Structure
app.py
chatbot.py
rag_pipeline.py
weaviate_server.py
requirements.txt
README.md
Running Locally
Install dependencies:
pip install -r requirements.txt
Configure the required environment variables:
- COHERE_API_KEY
- HF_TOKEN
- WEAVIATE_URL (if applicable)
- WEAVIATE_API_KEY (if applicable)
Launch the application:
python app.py
System Architecture
Deployment
The project is designed for deployment using:
- Hugging Face Models
- ALLAM language model
- Weaviate Vector Database
- Cohere Language model
Demo
Future Improvements
- Multi-turn conversation memory
- Expanded healthcare knowledge base
Author
Shahad Aljohani
Recent Computer Science Undergraduate
Privacy Notice
This repository contains a public research version. Private implementation details, API keys, deployment configurations, and some internal pipeline components are excluded.
The complete implementation is maintained privately.