Client-Side Fraud Detection Model for Data Security
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional software solutions fail to efficiently identify and combat fraudulent data entry, particularly in large datasets, due to resource-intensive encryption methods and the difficulty in verifying customer information across multiple networks and computing infrastructures.
Innovation Solution
A system and method that uses a server to generate a network page with input fields, monitor client devices, and apply a fraud detection model to determine the likelihood of fraud, prompting users to modify inputs and transmit electronic profiles for verification, thereby enhancing data security and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional encryption methods are used to secure data transmission, then data security is improved, but computing resources and processing time are significantly increased
Solution Approach 1:
The system performs preliminary fraud detection analysis on client devices before data is transmitted to the server. A fraud detection model is executed locally on the client device to identify suspicious patterns in real-time, preventing resource-intensive server-side encryption from being applied to obviously fraudulent data. This preliminary action filters out high-risk inputs before they consume significant computing resources.
Solution Approach 2:
The patent implements different security measures based on the local quality or risk level of each data input. Instead of uniformly applying heavy encryption to all data, the system dynamically adjusts security intensity based on fraud risk assessment. Low-risk data receives minimal processing, while high-risk data triggers more rigorous verification, optimizing the balance between security and resource consumption.
2Measurement precision
If manual verification of customer information is performed, then accuracy of fraud detection is improved, but time consumption and operational complexity increase
Solution Approach 1:
The patent replaces manual mechanical verification processes with an automated fraud detection model that runs on client devices. The model uses machine learning algorithms to automatically analyze input data patterns, device characteristics, and behavioral metrics, substituting human reviewers with an automated system that processes data faster and with consistent accuracy.
Solution Approach 2:
The fraud detection system performs self-service by automatically executing the fraud detection model on the client device without requiring manual intervention. The system independently collects device information, analyzes input patterns, generates fraud risk assessments, and triggers appropriate verification workflows autonomously, eliminating the need for manual review of low-risk cases.
3Difficulty of detecting and measuring
If comprehensive monitoring of input devices is implemented, then fraud detection capability is improved, but device complexity and data processing requirements increase
Solution Approach 1:
The patent extracts only the essential monitoring functions needed for fraud detection from the complete set of possible input device capabilities. Instead of monitoring all device functions comprehensively, the system selectively extracts and monitors specific input patterns, device identifiers, and behavioral metrics that are most indicative of fraud, reducing complexity while maintaining detection effectiveness.
Solution Approach 2:
The fraud detection model serves multiple functions simultaneously: it analyzes input data validity, assesses device risk profiles, detects fraudulent patterns, and triggers verification workflows. This multi-functional approach consolidates what would otherwise require separate monitoring systems into a single universal fraud detection engine, reducing overall system complexity.
Data Source
AI summary
Disclosed method comprises generating a fraud detection model comprising an algorithm to determine a likelihood of fraud for a client response based on data from similar customers responding to similar questions and data generated from tracking the client computing device. In response to the likelihood of fraud satisfying a threshold, the method comprises querying and displaying an electronic profile of the client, and inquiring the client as to whether the client is willing to modify any of the responses while monitoring the client device. The method comprises transmitting a fraud value to the customer database, in response to receiving the same input, form the client computing device.