🌐 ASTITVA

Academic Support Tool For Intervention, Tracking & Value Alignment
📍 Built for SNU-SIH Hackathon

Please do read along to understand deeply about our project, cause and solution! :)

We focused on being relevant to the problem statement, not over engineering it, because our solution has to be scalable and easy to use! (more about it below)

📄 Problem Statement ---

🖼️ Project Snapshots

🎯 Main Hackathon Theme

Sample Excel Sheet - Google Sheets Link

🔔 Notification System

🤖 ML Model (99% Accuracy)

⚙️ Backend Exploration

(Started with Java → finalized Node.js)

---

🚩 Problem Statement

By the time term-end marks reveal failures, struggling students have often disengaged beyond recovery, which often result in Dropouts.

• Attendance is in one spreadsheet.
• Test scores in another.
• Fee-payment data in a third.

👉 No single view exists to show when a student is slipping in multiple areas simultaneously.

Plus the problem statement says that Commercial analytics (Such as blackboard itself) platforms are expensive and demand heavy maintenance—beyond the reach of most public institutes.

✅ That’s exactly where we came with the idea of ASTITVA — a low-cost, scalable, and practical solution fully aligned with the problem statement.

---

ps - We would like to pursue this idea deeply into the "Internet in a box" concept in our next level, so we can reach the remotest of the village, and even without internet connection, the teachers and admins can care and analyse the students. That will be our implementation plan in the later stages!

---

💡 Our Solution — ASTITVA

A lightweight, transparent, and ML-powered platform that:

1. 📂 Ingests spreadsheets (attendance, test results, fees, etc.).
2. ⚡ Applies ML + rule-based thresholds to detect at-risk students (dropout likelihood).
3. 🎨 Visualizes risks in a clean dashboard:
   • ✅ Green (0–75% risk)
   • ⚠️ Orange (75–90% risk)
   • 🔴 Red (90%+ risk)
4. 📱 Notifies mentors, guardians, and students via SMS & Email.
5. 👨‍🏫 Provides dual roles:
   • Admin View → consolidated risk dashboard of all students.
   • Student View → personal profile + weak areas & tips.
6. 🛠️ Low-cost, easy-to-use, no vendor lock-in.

---

📝 Note for Judges

During development, we explored:

• 🤖 AI-based mentorship chatbots
• 📊 Complex deep learning models
• ☁️ Enterprise-grade cloud dashboards

But we chose to stay true to the problem statement.
We kept reminding ourselves that the govt. already knows costly platforms exist — what’s missing is a viable, grassroots-friendly system.

👉 ASTITVA directly solves every pain-point in the problem statement while staying lightweight, practical, and affordable.

---

🏗️ Features at a Glance

✅ Student Risk Prediction (ML + rule-based logic)
✅ Excel Upload (drag & drop interface)

✅ Dashboard with filters, sorting, and color-coded risk levels

✅ Student Profiles (auto-filled from uploaded Excel)

✅ Regular Notifications (Email & SMS)

✅ Easy to understand Interface

✅ Download RISK wise / Attendance <75% wise Excel sheets (since many still prefer Excel workflows)

✅ Lightweight, scalable, and easy to deploy

---

🔧 Tech Stack

• Frontend: React (Vite, TypeScript, shadcn-ui, Tailwind CSS)
• Backend: Node.js (Express)
• Database: MongoDB
• ML Model: Scikit-learn (Python, 99% accuracy)
• Notifications: Twilio (SMS), Nodemailer (Email)

---

🚀 Getting Started

Prerequisites

• Node.js & npm installed → nvm install guide
• Supabase (local for the future, cloud currently)

Run Locally

# Step 1: Clone the repository
git clone <YOUR_GIT_URL>
cd <YOUR_PROJECT_NAME>

# Step 2: Install dependencies
npm i

# Step 3: Start development server
npm run dev