Trust Platform Using ML for Card Location Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current merchant applications lack effective verification of user credibility, leading to fraudulent transactions and inefficient use of resources in addressing such issues, as they cannot reliably authenticate users or verify their locations.
Innovation Solution
A trust platform utilizing machine learning and trusted transaction card locations to generate a geographical map of trusted transaction cards, determining trust scores and identifying networks of trusted cards, which enhances security and reduces fraudulent activities by authenticating users and conserving resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional merchant applications are used without trust verification, then ease of operation is maintained, but reliability of transactions deteriorates due to fraudulent activities
Solution Approach 1:
The patent introduces a trust platform as an intermediary system between merchants and users. This platform uses machine learning models to assess user credibility and determine trust scores, acting as a mediator that verifies user identity and location without requiring complex verification procedures at the merchant端. The trust platform generates geographical maps of trusted users, enabling secure transactions while maintaining ease of operation for end users.
2Reliability
If comprehensive trust verification is implemented, then reliability of transaction authentication is improved, but loss of computing resources increases due to processing large amounts of user data
Solution Approach 1:
The trust platform performs preliminary trust assessments by analyzing user data, device information, and location data before transactions occur. Machine learning models pre-compute trust scores and generate geographical maps of trusted users in advance. This preliminary action allows the system to quickly verify user credibility during transactions without processing large amounts of data in real-time, thereby reducing computing resource consumption while maintaining high authentication accuracy.
3Reliability
If location-based trust verification is used, then reliability of user identity confirmation is improved, but measurement precision requirements increase for location data
Solution Approach 1:
The trust platform uses location data with varying precision requirements based on the specific transaction context. For low-risk transactions, approximate location matching within a geographical zone suffices. For high-risk transactions, the system can require more precise location verification. The machine learning model adjusts the precision requirement dynamically, using partial verification when sufficient and excessive verification only when necessary, thereby balancing reliability with measurement precision requirements.
Data Source
AI summary
A device may receive, from client devices of users, user data identifying the users, client device data identifying the client devices, and transaction card data identifying transaction cards, and may receive transaction account data identifying transaction accounts. The device may process the user data, the client device data, the transaction card data, and the transaction account data, with a machine learning model, to determine trust scores for the transaction cards, and may identify trusted transaction cards based on the trust scores. The device may receive, from trusted client devices associated with the trusted transaction cards, location data identifying locations of the trusted client devices and communication data indicating communications between the trusted transaction cards and the trusted client devices. The device may generate a card mapping for the trusted transaction cards based on the location data and the communication data, and may perform actions based on the card mapping.