Artificial Intelligence-Enabled Computational Techniques in Determining the Use Behavior of Crowdfunding Platforms

Crowdfunding has garnered much attention from academics and practitioners as a practical approach to raising funds for projects or ventures. This research aims to understand the use of Artificial Intelligence (AI) algorithms in determining the use of Shariah-compliant Peer-to-Business (P2B) crowdfunding platforms. This study adopts Technology Acceptance Model and Unified Theory of Use of Technology to establish the conceptual model. Data is collected from 116 micro, small, and medium enterprises (MSMEs). Seven state-of-the-art AI algorithms, including Decision Tree (DT), Extra Tree (ET), K-Nearest Neighbors (KNN), Lo-gistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost), are applied to evaluate the essential features of the Actual Use Behavior (AUB). The results indicate that RF, ET, and KNN are the top three methods in all AUBs. The findings suggest that incorporating AI in Shariah-compliant P2B crowdfunding enhances financial inclusivity by enabling more efficient resource allocation and supporting the growth of MSMEs