Risk Score Recommendation Engine for Exam Fraud Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Online and testing center-based candidate evaluation systems face challenges in preventing cheating and fraud, as individuals or organizations attempt to circumvent security measures, necessitating a system to monitor and identify suspicious activities effectively.

Innovation Solution

A distributed computing environment with a data store server, model server, and resource management server that generates risk scores for candidates, test centers, and proctors using machine learning models, triggering predefined actions when aggregate risk scores exceed a threshold, such as live video feed monitoring or enhanced identity verification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If online and testing center-based candidate evaluation systems are implemented, then convenience and efficiency are improved, but vulnerability to cheating and fraud increases

Engineering Contradiction:

Improveevaluation efficiencyVSAvoidcheating and fraud risk

Core Design Contradiction:

ProductivityVSObject-affected harmful factors

Solution Approach 1:

The system performs preliminary risk assessment by generating risk scores for candidates, test centers, and proctors before the examination process begins. Machine learning models analyze historical data and entity attributes to predict potential fraud risks in advance, enabling preventive measures to be taken before cheating can occur.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors examination activities and provides real-time feedback by updating risk scores based on observed behaviors. When suspicious activities are detected during the examination, the system generates alerts and triggers investigative actions, creating a closed-loop feedback mechanism that adapts to emerging threats.

Inventive Principle:

Principle #23Feedback

2Difficulty of detecting and measuring

If risk scoring and monitoring systems are implemented, then detection capability is improved, but system complexity increases

Engineering Contradiction:

Improvefraud detection capabilityVSAvoidsystem architecture complexity

Core Design Contradiction:

Difficulty of detecting and measuringVSDevice complexity

Solution Approach 1:

The system divides the fraud detection task into multiple independent machine learning models, each responsible for scoring specific entities (candidates, test centers, proctors). This segmentation allows each model to focus on specific fraud patterns and reduces the overall computational complexity compared to a single monolithic detection system.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an aggregate risk score as an intermediary metric that synthesizes individual entity risk scores. This aggregate score serves as a simplified decision indicator for triggering investigative actions, mediating between complex individual risk assessments and straightforward fraud detection decisions.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Measurement precision

If real-time monitoring and aggregate risk scoring are implemented, then fraud identification accuracy is improved, but processing time increases

Engineering Contradiction:

Improvefraud detection accuracyVSAvoidexamination processing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system calculates risk scores for all entities in advance and stores them, then only performs aggregate risk calculations and detailed analysis when thresholds are exceeded. This partial action approach computes full risk assessments only when necessary, reducing overall processing time while maintaining detection accuracy for suspicious cases.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

Individual entity risk scores are pre-calculated and stored before the examination process begins. This preliminary computation of base risk scores eliminates the need to recalculate them in real-time, allowing the system to quickly aggregate scores and identify fraudulent activities without significant processing delays.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12175549B2Recommendation engine for testing conditions based on evaluation of test entity scores

Publication Date: 2024.12.24 NCS PEARSON INC

AI summary

Systems and methods may involve processing of entity data by machine learning models to produce one or more entity and/aggregate risk scores and/or aggregate anticipated risk scores, which may be compared to one or more thresholds to determine when one or more predefined actions should be taken. The entity data may be collected for various entities related to an exam registration and delivery process, which may include a candidate, an exam, a test center, an exam registration event, a proctor, and an exam delivery event. The exam registration and delivery process may include multiple states—each being associated with a different set of entities. Aggregate risk scores for a given state may be calculated using only entity data for the set of entities associated with that state. The predetermined actions taken may also be dependent on the current state.