Abstract

The Classification of digital wallet payment applications based on transaction intensity is crucial for understanding user behaviour and optimising digital Financial Services. The study the application of the machine learning techniques specifically Decision tree, Random forest, Gradient Boosting to classify UPI applications into ‘High' and ‘Low' transaction' categories. The performance of each application is evaluated using Accuracy, Precision, Recall and F1 scores. It demonstrates the effectiveness of UPI app classification AI model classification. The feature importance analysis also revealed that Total Transaction value (Vtt)and Customer-initiated Transactions (Ctn)were most significant indicators.