Biometric Authentication Score Calculation for Transaction Authorization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems lack the capability to effectively combine smartcard technology with biometric authentication for transaction authorization, requiring a method to verify user identity using both policy rulesets and biometric data for secure transaction processing.
Innovation Solution
A computer-implemented method and system that receives a policy message with a ruleset and biometric parameters, calculates an authentication score using machine learning, and determines transaction authorization based on the score, transmitting a decision message to indicate whether the transaction satisfies the policy ruleset.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If biometric authentication is combined with smartcard policy ruleset verification, then transaction security is improved, but system complexity increases
Solution Approach 1:
The patent combines biometric authentication with smartcard policy ruleset verification into a unified authorization system. The authorization server integrates both biometric score calculation and policy ruleset evaluation, merging previously separate authentication and authorization functions into a single coordinated process that enhances security while managing complexity through integration.
Solution Approach 2:
The authorization server acts as an intermediary between the biometric measurement system and the policy ruleset database. It receives biometric data, calculates authentication scores, queries the policy ruleset, and synthesizes both inputs to generate authorization decisions, mediating between different security layers without requiring direct integration between biometric systems and policy databases.
2Measurement precision
If machine learning algorithms are used to calculate authentication scores, then authentication accuracy is improved, but processing time increases
Solution Approach 1:
The system applies machine learning algorithms selectively based on risk assessment. For low-risk transactions, simpler authentication methods may suffice, while high-risk transactions trigger the more computationally intensive machine learning-based biometric analysis. This partial application of complex processing reduces average processing time while maintaining high authentication accuracy when needed.
Solution Approach 2:
Biometric templates and machine learning models are pre-trained and stored in the authorization server before actual transactions occur. During transaction processing, the system performs rapid matching against pre-computed models rather than performing full machine learning inference from raw data, significantly reducing real-time processing requirements while maintaining authentication accuracy.
3Reliability
If biometric data is processed through policy ruleset, then authorization control is improved, but computational load increases
Solution Approach 1:
The authorization process is segmented into distinct stages: biometric data collection, authentication score calculation, policy ruleset querying, and final decision generation. Each stage processes only the necessary data for its specific function, avoiding redundant computations. The policy ruleset is segmented into separate queryable rules that can be efficiently evaluated without processing entire rulebases unnecessarily.
Solution Approach 2:
The patent extracts only the essential biometric features needed for authentication rather than processing complete biometric datasets. The authorization server extracts relevant authentication scores from biometric measurements and only queries policy ruleset elements necessary for the specific transaction context, reducing computational load by eliminating unnecessary data processing while maintaining authorization control.
Data Source
AI summary
Disclosed are methods for authorizing a transaction, including receiving a policy message, the policy message including a policy ruleset for determining whether a transaction is authorized and biometric parameters for a machine learning algorithm for authenticating an identity of a user involved in a transaction, receiving biometric measurement data associated with a biometric measurement of a user involved in the transaction, calculating an authentication score based on the biometric measurement data using the machine learning algorithm, wherein the authentication score includes an indication of whether an identity of the user is authenticated based on the biometric measurement data, determining whether the transaction satisfies the policy ruleset for determining authorization of the transaction, and transmitting a decision message, wherein the decision message includes an indication of whether the transaction satisfies the policy ruleset. Systems and computer program products are also disclosed.