Multi-dimensional Device ID for Real-time Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for preventing online fraud in electronic commerce are inadequate, as they rely heavily on consumer vigilance and are difficult to manage, and existing technologies struggle to uniquely identify devices and users amidst rapid changes and updates, leading to high costs and inefficiencies in fraud detection and recovery.
Innovation Solution
A centralized webserver service that collects and analyzes comprehensive dossiers of user device configurations and behaviors in real-time, using multi-layer behavioral device identification and Smart-Agents to create virtual agents that learn user behavior patterns, providing a fraud score to assist websites in transaction decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If comprehensive device fingerprinting and behavioral analysis are implemented to uniquely identify users, then fraud detection accuracy is improved, but system complexity and data processing requirements increase
Solution Approach 1:
The system segments device identification into multiple independent fingerprinting dimensions (hardware attributes, software attributes, behavioral attributes) that can be collected and analyzed separately, then combined to form a comprehensive device profile. This segmentation allows the system to manage complexity by processing individual attribute sets rather than overwhelming monolithic data structures.
Solution Approach 2:
The patent introduces behavioral dimensions (click patterns, navigation sequences, time spent on pages) alongside traditional device fingerprinting dimensions (hardware IDs, screen resolution, browser version). By adding these temporal and interactive dimensions, the system achieves more precise fraud detection without relying solely on static device attributes that can be easily spoofed.
2Reliability
If real-time behavioral analysis is performed on all user transactions, then fraud prevention effectiveness is improved, but processing time and computational resources increase
Solution Approach 1:
The system performs preliminary device fingerprinting and baseline behavioral analysis during initial user interactions and off-peak periods, building device profiles and establishing normal behavior patterns before fraud incidents occur. This preliminary characterization allows real-time transaction monitoring to focus on detecting deviations from established baselines rather than analyzing all attributes from scratch for each transaction.
Solution Approach 2:
The system applies full behavioral analysis selectively to high-risk transactions based on initial risk scoring, while using lighter-weight monitoring for low-risk transactions. This partial application of comprehensive analysis reduces overall processing overhead while maintaining high fraud detection effectiveness by concentrating resources on suspicious activities.
3Measurement precision
If multiple device identifiers and behavioral descriptors are collected, then user identification uniqueness is improved, but data collection privacy concerns and user resistance increase
Solution Approach 1:
The system collects device fingerprinting data passively through automatically executed JavaScript code that gathers device attributes without requiring explicit user input or consent for each data point. Users inadvertently provide behavioral data through their natural interaction patterns (clicks, scrolls, navigation) while using the website, eliminating the need for cumbersome authentication forms or privacy- intrusive permission requests.
Solution Approach 2:
The patent employs an intermediary JavaScript agent that runs in the user's browser and acts as a mediator between the website and the user's device. This intermediary collects device attributes and behavioral data locally, then transmits aggregated fingerprints to the server, providing a layer of abstraction that protects user privacy while enabling comprehensive device identification.
4Reliability
If centralized server infrastructure is used to store and analyze device dossiers, then fraud detection capability is improved, but system vulnerability to attacks and single points of failure increase
Solution Approach 1:
The system segments the centralized server infrastructure into multiple distributed server nodes that collectively store and process device fingerprinting data. No single server holds the complete dataset, and servers can operate independently, reducing the impact of attacks on individual nodes and eliminating single points of failure while maintaining centralized coordination for fraud detection logic.
Data Source
AI summary
A real-time fraud prevention system enables merchants and commercial organizations on-line to assess and protect themselves from high-risk users. A centralized database is configured to build and store dossiers of user devices and behaviors collected from subscriber websites in real-time. Real, low-risk users have webpage click navigation behaviors that are assumed to be very different than those of fraudsters. Individual user devices are distinguished from others by hundreds of points of user-device configuration data each independently maintains. A client agent provokes user devices to volunteer configuration data when a user visits respective webpages at independent websites. A collection of comprehensive dossiers of user devices is organized by their identifying information, and used calculating a fraud score in real-time. Each corresponding website is thereby assisted in deciding whether to allow a proposed transaction to be concluded with the particular user and their device.