Title: Medical crowdfunding in India during COVID-19: predictive modelling of campaign strength using XGBoost and random forest

Authors: Rangapriya Saivasan; Madhavi Lokhande

Addresses: International School of Management Excellence, Research Centre, Affiliated to the University of Mysore, India ' International School of Management Excellence, Research Centre, Affiliated to the University of Mysore, India

Abstract: Medical crowd funding emerged as an important channel, during COVID-19, to bridge the chasm between the overburdened public healthcare system and expensive private medical services. In India, the percentage of funds raised by such campaigns vis-à-vis the targeted amount is low in most cases. This study identifies the variables that have an impact on the performance of the campaign and leverages them to predict the 'campaign strength'. A sample of 305 COVID-related campaigns is analysed using descriptive statistics and text analytics. The strength of the campaigns is classified as - weak, moderate and strong based on their success rate (fundraised/target amount). Machine learning algorithms are used to predict the strength of the campaign. XGBoost predicts the campaign strength with 74.19% accuracy while Random Forest predicts campaign strength with 75% accuracy. The insights can be used to improve the layout design of the platforms, achieve better campaign strength and enhance stakeholders' experience.

Keywords: medical crowdfunding; predictive modelling; COVID-19; fundraising; machine learning; XGBoost; random forest.

DOI: 10.1504/IJBIR.2025.150771

International Journal of Business Innovation and Research, 2025 Vol.38 No.4, pp.505 - 529

Received: 23 Nov 2021
Accepted: 23 Oct 2022
Published online: 23 Dec 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article