✨ A Smart India Hackathon (SIH) 2024 project — Machine Learning module for Sangam: A platform that simplifies interdepartmental collaboration in government.
Sangam is a collaborative digital platform built to empower interdepartmental communication and cooperation for government departments such as Water Supply, Gas, Electricity, Public Works, and more.
Key objectives of Sangam:
- 👥 Foster collaboration between departments on shared project sites
- 🛠️ Facilitate joint project planning and task scheduling
- 📊 Enable real-time monitoring and transparency
- 🤝 Break silos in urban governance & infrastructure delivery
The ML module in this repo powers smart insights, predictions, and anomaly detection to drive data-driven decisions.
This repository contains all ML-related models and scripts that support the Sangam platform by enabling predictive intelligence.
-
📉 Anomaly Detection Model
Detects irregularities in task updates, cost flow, and resource usage — helping departments flag suspicious or inconsistent behavior early. -
💸 Cost Optimization Prediction Model
Predicts potential cost reduction strategies by analyzing historical project data, resource allocation patterns, and timelines. -
⏳ Task Completion Status Predictor
Uses machine learning to forecast whether a task/project is likely to meet its deadline or not — allowing proactive interventions.
🔗 Live Deployment Link: Click here 🔗 Render Deployment Link: Click here
Getting started
## 🔧 Installation
git clone https://github.com/anjaliy11/Sangam_ML.git
cd Sangam_ML
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
pip install -r requirements.txt