Behavior-Based User Account Classification for Remote Payment Onboarding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional payment service systems face challenges in accurately identifying and classifying user accounts, especially during remote onboarding, due to insufficient information provided by users, which complicates the verification of user identities and authentication, particularly for younger users lacking identification documents and credit history, making it difficult to detect fraud.
Innovation Solution
Utilizing AI models trained on contextual data from user behavior to classify user accounts into different types, such as age ranges, and selectively running these models to conserve resources, while dynamically tailoring application interfaces and functionality based on user types, and implementing automated fraud detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional authentication methods are used during remote onboarding, then user identity verification can be performed, but younger users lacking identification documents and credit history cannot be properly authenticated
Solution Approach 1:
The system changes the parameters used for authentication from traditional static documents (ID, credit history) to dynamic behavioral parameters (spending patterns, transaction frequency, device usage). This allows the authentication system to adapt to younger users who lack conventional identification documents but exhibit recognizable behavioral patterns.
Solution Approach 2:
The patent replaces the mechanical system of document-based verification with an AI-driven behavioral analysis system. Instead of physically examining identification documents, the system uses machine learning models to analyze digital footprints and behavioral data, substituting one verification mechanism with a more adaptable digital approach.
2Measurement precision
If AI models are continuously run to classify all user accounts, then user classification accuracy is improved, but computing resources are wasted
Solution Approach 1:
Instead of applying AI classification to all user accounts continuously, the system applies partial action by selectively running models only for users who meet specific criteria (new users, unusual behavior patterns, or those requiring verification). This reduces unnecessary computational overhead while maintaining classification accuracy where needed.
Solution Approach 2:
The system implements feedback mechanisms where classification results and resource usage metrics are continuously monitored. Based on this feedback, the system dynamically adjusts when and how AI models are deployed, optimizing the balance between classification accuracy and resource consumption through iterative improvement.
3Ease of operation
If generic user interfaces are provided to all users, then system simplicity is maintained, but user experience is not optimized for different user types
Solution Approach 1:
The system applies local quality by providing different interface characteristics to different user segments. Instead of uniformly simplifying or complicating the interface for all users, it tailors specific interface elements (such as guidance levels, feature visibility, or interaction patterns) to match the needs of specific user types like younger users versus experienced users.
Solution Approach 2:
The user interface transitions from a static generic design to a dynamic adaptive design that automatically adjusts its properties based on the user's classified type. The interface can dynamically modify its complexity, information density, and interaction patterns in real-time based on user behavior and classification, maintaining simplicity where needed while providing advanced features where appropriate.
Data Source
AI summary
Determining user types from behavior is described. An artificial intelligence (AI) model is trained to classify user accounts of a payment service into different user types using contextual data associated with processed payments between the user accounts. The AI model is used to analyze additional contextual data associated with additional payments between additional user accounts to classify the additional user accounts, and, if a particular user account of the additional user accounts is associated with a user type of the different user types that requires an action to be performed, an instruction is sent to a user device associated with the particular user account to cause a payment application to present a user interface element prompting a user to perform the action, and, based on whether the action was performed, account data indicating whether the particular user account is an authorized account is stored in a datastore.