• Vinay chauhan
• prashant mann

The increasing use of unified payment interfaces ( upi ) for online payments has provided the convenience and efficiency synonymous with digital transactions, but this has resulted in alarmingly high rates of fraud, putting the system's reliability and stability at risk. This paper explores how to solve the upi fraud detection problem by using a hmm Markov model. The multichannel upi communications scheme is investigated and modelled by using the hmm framework. This accounting system is similar to a typical sherlock Holmes system, in that it is able to distinguish between legitimate and fraudulent businesses. Here's how it works: the hmm is first introduced to the cardholder by analysing their previous transaction behaviour. This involves comparing sales volumes, locations, frequency, and other patterns. When a cardholder initiates a new upi transaction, the trained hmm learns to recognize what normal '' behaviour looks like for that particular cardholder. If the transaction is judged to be valid by hmm to correspond to the cardholder's specific transaction, it is deemed valid. However, if the hmm raises a red flag and indicates that the behaviour does not conform to scholarly principles, it is considered fraudulent, prompting a further investigation. This project aims to show how data modelling and machine learning can be used to shield individuals and financial institutions from upi fraud. We're making a significant leap forward in terms of not only convenience but also safety in upi transactions by implementing a system that can detect suspicious activity in real time, while still maintaining that they are all digital. Keywords: fraud detection, unified payment interface ( upi), and hidden.)

"UPI fraud detection using ml", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 4, page no.k53-k59, April-2024, Available :http://www.jetir.org/papers/JETIR2404A10.pdf

"UPI fraud detection using ml", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 4, page no. ppk53-k59, April-2024, Available at : http://www.jetir.org/papers/JETIR2404A10.pdf