Recruitment Process Graphs for Unsupervised Fraud and Collusion Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing recruitment processes face challenges in detecting fraud, collusion between teams/agencies, and biased treatment of candidates, which are technically difficult to identify and address.
Innovation Solution
A recruitment process graph-based unsupervised anomaly detection system generates knowledge graphs from recruitment logs, uses graph embeddings trained to distinguish between genuine and fraudulent candidates, and employs clustering to identify anomalies, enabling detection of fraud and collusion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional recruitment credential analysis is used, then the recruitment process is simple and fast, but fraud and collusion cannot be effectively detected
Solution Approach 1:
The patent transforms the recruitment analysis from traditional credential verification into a multi-dimensional graph structure where candidates, recruiters, and interactions are nodes and edges. This dimensional transformation enables detection of complex fraud patterns and collusion that cannot be identified through linear credential checking alone.
Solution Approach 2:
The patent replaces traditional mechanical credential verification methods with graph embedding techniques and machine learning algorithms. The system uses unsupervised anomaly detection on graph embeddings to automatically identify fraudulent patterns, substituting manual or rule-based verification with intelligent computational methods.
2Reliability
If graph embedding and clustering methods are implemented, then fraud and collusion can be detected, but the processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary actions by pre-computing graph embeddings and organizing recruitment data into structured knowledge graphs before anomaly detection is needed. This preprocessing enables faster real-time detection when actual recruitment decisions are being made, as the heavy computational work is done in advance.
Solution Approach 2:
The patent changes parameters by transforming recruitment data into graph embedding representations with specific dimensional properties. By adjusting embedding dimensions and clustering parameters, the system optimizes the balance between detection accuracy and computational efficiency, enabling scalable fraud detection.
Data Source
AI summary
In some examples, recruitment process graph based unsupervised anomaly detection may include obtaining log data associated with a recruitment process for a plurality of candidates, and generating knowledge graphs and graph embeddings. The graph embeddings may be trained to include a plurality of properties such that graph embeddings of genuine candidate hires and fraudulent candidate hires are appropriately spaced in a vector space. The trained graph embeddings may be clustered to generate a plurality of embedding clusters that include a genuine candidate cluster, and a fraudulent candidate cluster. For a new candidate graph embedding for a new candidate, a determination may be made as to whether the new candidate graph embedding belongs to the genuine candidate cluster, to the fraudulent candidate cluster, or to an anomalous cluster, and instructions may be generated to respectively retain or suspend the new candidate.