|
WAPDA MDC AI Assistant is a professional-grade, bilingual AI console built to help field staff and consumers navigate Meter Data Control (MDC) and WAPDA operations instantly β day or night, in whichever language they think in. Developed by Ayesha Javid, it replaces slow manual lookups with a single conversational interface that understands English, Urdu script, and Roman Urdu equally well.
|
|
| Category | What It Does |
|---|---|
| π§ Intelligent Core |
|
| π Multi-Lingual Engine |
|
| π¨ Cinematic UI |
|
| πΎ Persistence |
|
flowchart TD
A[π€ User types a question] --> B{π Language Detected}
B -->|English| C[NLP Match Engine]
B -->|Urdu Script| C
B -->|Roman Urdu| C
C --> D{π Match found in 160+ Q&A?}
D -->|β
Yes| E[β‘ Instant Answer Returned]
D -->|β No| F[π¨ Smart Filter: Irrelevant Alert]
E --> G[πΎ Saved to Session Memory]
F --> G
G --> H[π Recents Sidebar Updated]
| Language | Example Query | Understood? |
|---|---|---|
| π¬π§ English | "What is MDC in WAPDA billing?" | β |
| π΅π° Urdu (Script) | "Ω ΫΩΉΨ± Ψ±ΫΪΩΪ― Ϊ©ΫΨ³Ϋ ΪΫΪ© Ϊ©Ψ±ΫΪΊΨ" | β |
| π£οΈ Roman Urdu | "Bill ziyada q aya hai?" | β |
| β Off-topic | "What's the weather today?" | π¨ Flagged as irrelevant |
| Layer | Technology | Role |
|---|---|---|
| βοΈ Backend | Python 3.9 & Flask | API logic, query matching, session control |
| π¨ Frontend | Vanilla HTML5, CSS3 (Glassmorphism), JavaScript (ES6+) | Cinematic chat console UI |
| πΎ Session Layer | Flask-Session | Per-user chat memory (last 20 messages) |
| βοΈ Deployment | Hugging Face Spaces (Dockerized) | Live, always-on hosting |
π Try it Live β No Setup Needed π
π» Click to expand: Run it locally
1. Clone the repository
git clone https://github.com/ayeshajavid91-star/wapda-chatbot.git
cd wapda-chatbot2. Install dependencies
pip install -r requirements.txt3. Run the console
python app.py4. Open in browser
http://localhost:5000
π³ Click to expand: Run with Docker
docker build -t wapda-mdc-assistant .
docker run -p 5000:5000 wapda-mdc-assistantwapda-chatbot/
β
βββ βοΈ app.py β Core AI logic & API endpoints
βββ π¨ templates/
β βββ index.html β Enterprise chat console UI
βββ π qa_database.json β 160+ question knowledge base
βββ π¦ requirements.txt β System dependencies
βββ π³ Dockerfile β Hugging Face deployment config
βββ π README.md
python flask chatbot ai-assistant nlp bilingual-chatbot
- 160+ Q&A knowledge base
- English / Urdu / Roman Urdu support
- Smart irrelevant-query filter
- Session memory (last 20 messages)
- Voice input for field staff
- Admin panel to expand the Q&A database
- SMS/WhatsApp integration for offline consumers
All Rights Reserved.
This project and its source code are the intellectual property of the author. No part of this repository β including the code, design, or documentation β may be copied, modified, distributed, used, or reproduced in any form without the explicit written permission of the author.
Β© 2026 Ayesha Javid. Unauthorized use is strictly prohibited.
For permissions or licensing inquiries, contact: ayeshajavid91@gmail.com
