Multi-Turn AI Chatbot With using Llama3

 (Artifical Intelligence / Natural language Processing / Conversational AI)

Project IDEA:

• We first install ollama Software and then run llama3 model locally, Then llama3 model exposes it's api then we use this api in our website. We build backend & frontend In backend we do query classification and validation about query length and taking input feedback after 4 to 5 queries and storing the queries in database with validation. Then making two roles User & Admin. Admin see feedback analytics graph, statics of multi-dimmensional query classification and see response time from starting to end

Phases:

• The project has already divided into three phases
  1. Phase 1: Chatbot System Development (Frontend + Backend)
  2. Phase 2: Chat Execution + Data Storage System
  3. Phase 3: Analytics System

Phase 1: ChatBot System Development (Frontend + Backend)

FR1: System Initialization

• LLaMA 3 local model loading
• Backend server initialization
• Database connection setup
• Analytics module initialization

FR2: Frontend Chat Interface Development

• Google OAuth login page
• Chat interface (user input + response display)
• Feedback panel (rating, correctness, length type)

FR3: Backend Server Development

• The system shall implement a backend using Flask/FastAPI.
• The backend shall handle:
• User authentication
• Session creation and management
• Message routing to model
• Data storage operations

FR4: LLaMA 3 Model Integration

• The system shall integrate LLaMA 3 as the core chatbot engine.
• The system shall process user queries locally through the model.

FR5: Service Connectivity

• The system shall connect:
  1. Frontend UI
  2. Backend API
  3. LLaMA 3 model inference engine
  4. Database
  5. Analytics module