Portable Financial Device Recommendation via ML Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems for recommending portable financial devices for payment transactions lack accuracy in detecting geolocation, require customers to carry multiple devices, and fail to consider real-time data, making them unreliable for maximizing reward benefits.
Innovation Solution
A computing system that establishes secure communication with external APIs to fetch real-time banking data, determines transactional parameters, and uses a Machine Learning-based model to generate an optimal score for each device, recommending the best device for a payment transaction based on these parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If geolocation information is used to recommend portable financial devices, then the recommendation process is simplified, but the accuracy of device detection is poor
Solution Approach 1:
The patent introduces a third-party recommendation system as an intermediary between the customer and multiple portable financial devices. This system collects comprehensive data from various sources (banking accounts, transaction records, merchant information) and processes it through machine learning models to generate accurate device recommendations, eliminating the need for direct geolocation detection while improving recommendation precision
2Reliability
If customers carry multiple portable financial devices to access different reward programs, then they can maximize reward benefits, but the convenience and ease of use are reduced
Solution Approach 1:
The system enables automatic self-service by analyzing customer transaction patterns, account balances, and merchant information in real-time. The machine learning model automatically determines which portable financial device should be used for each transaction to maximize reward benefits, eliminating the need for customers to manually select from multiple devices and reducing the burden of carrying them
3Device complexity
If conventional systems recommend portable financial devices without considering real-time data, then the system complexity is reduced, but the reliability of recommendations is poor
Solution Approach 1:
The system performs preliminary data collection and processing by continuously gathering information from banking accounts, transaction records, and merchant databases before making recommendations. This pre-processing of real-time data allows the machine learning model to generate reliable recommendations without adding significant operational complexity during the actual transaction moment
Data Source
AI summary
A system and method for recommending portable financial device for a payment transaction is disclosed. The method includes establishing a secure communication session with one or more external APIs during a payment transaction stage and fetching data representative of banking accounts associated with a customer from the one or more external APIs. The method further includes determining one or more transactional parameters associated with the payment transaction stage and generating an optimal score for each of the one or more portable financial devices by using Machine Learning (ML) based transaction model. The method includes identifying best suitable portable financial device with maximum optimal score and recommending the identified best suitable portable financial device for completing the payment transaction stage based on the identification. Further, the method includes outputting the recommended portable financial device on a graphical user interface of one or more electronic devices associated with the customer.