DAANSETU A Machine Learning-based Platform for Enhancing NGO-Donor Interactions

A machine learning-based end-to-end donation management system, DAANSETU facilitates the connection of verified Non-Governmental Organizations (NGOs) with donors of surplus resources. DAANSETU addresses a long-standing socio-environmental challenge: the squandering of excess food, clothing, medication, and educational resources when millions of people cannot access such resources. The efficacy, accountability, and transparency of a donation system are greatly enhanced through DAANSETU, leveraging demand forecasting, matching, and tracking. This document presents an overview of all stages of the DAANSETU system development, starting with system design, coding, testing, and finally deployment. The system was created using Firebase Realtime Database as a backend, Android Studio as a frontend, and various python- based machine learning libraries such as Scikit-learn, XGBoost, and TensorFlow for analytics. During a pilot test conducted from March 3rd to 16th, 2025, with five local NGOs and twelve donor organizations throughout the Pune district, the implemented platform recorded 428 donation events with a 93Keywords: Digital Donation Platform, Machine Learning, NGO-Donor Matching, Resource Optimization, Real-Time Tracking, Firebase, Android, XGBoost, Sustainable Development.