AI-Based Micro-Services for Real-Time Analysis of Crowdfunding Campaigns Performance

Crowdfunding platforms offer interactive portals which are designed to motivate peoples to donate, simplify and automate the donation process. They focus on personalized strategies to improve the donors experience and keep them engaged over the long-term. This is enhanced by using AI-based solutions to analyze campaigns profiles and to predict donors behaviors. However, existing Machine Learning (ML) -based tools do not provide real-time data analytics to suggest optimal engagement techniques. They are often research-only and do not provide iterative feedback, real-time scoring and intelligent recommendations to engage donors and to retain their participation over time. In this context, we propose an AI-based interactive approach which offers real-time micro-services for the prediction and analysis of donors behaviors and crowdfunding campaigns performance. This is enhanced by dynamic indicators to predict campaign progress and retain donors engagement. The prediction is performed thanks to real-time training datasets which continuously learn from the online database of the Tunisian crowdfunding platform ‘Cha9a9a.tn’.