Behavioral Transaction Authentication Using Machine-Learned Questions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing user authentication methods, such as username and password, are inadequate for distinguishing authentic users from malicious attackers, as they can be easily guessed or circumvented, and alternative methods like transaction-based questions are difficult for users and vulnerable to data capture by attackers.
Innovation Solution
A system that generates customized authentication questions based on user transaction patterns detected through machine learning, using clustering algorithms and neural networks to analyze transaction data, creating questions that only an authentic user would know.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional username and password authentication is used, then the authentication process is simple and fast, but the security is weak and easily circumvented by malicious attackers
Solution Approach 1:
The patent changes the authentication parameters from static credentials (username/password) to dynamic behavioral patterns (transaction timing, frequency, amount ranges). The system analyzes multiple transaction parameters and generates authentication questions based on detected patterns, making authentication both secure and user-friendly by leveraging natural user behavior rather than requiring memorized secrets
2Reliability
If transaction-based authentication questions are used, then the security is improved, but the questions are difficult for users and vulnerable to data capture by attackers
Solution Approach 1:
The system performs preliminary analysis of transaction data to detect and store user behavior patterns before authentication is needed. By pre-computing patterns from historical transactions and storing them as reference data, the system enables fast authentication questions to be generated without requiring users to recall complex transaction details, thus improving both security and user convenience
3Measurement precision
If machine learning algorithms are used to analyze transaction data, then the authentication accuracy is improved, but the system complexity increases
Solution Approach 1:
The patent extracts and isolates specific meaningful patterns from complex transaction data using machine learning algorithms. Instead of processing all raw transaction data directly during authentication, the system pre-extracts and stores simplified pattern representations (e.g., typical transaction times, frequency ranges, amount patterns). This extraction approach reduces the complexity of real-time authentication while maintaining high detection accuracy
Data Source
AI summary
Aspects described herein may allow for authenticating a user by generating a customized set of authentication questions based on patterns that are automatically detected and extracted from user data. The user data may include transaction data collected over a period of time. By automatically detecting user patterns that correspond to user behavior over a period of time, an authentication system may be able to generate information that is recognizable to an authentic user but difficult to guess or circumvent for any other user.