Vendor Email Fraud Detection Using Digital Communication Profiles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing security measures are ineffective in detecting and preventing external fraud, particularly those originating from compromised vendor accounts or impersonating vendors, as they rely on conventional defenses that do not examine vendor-specific information and fail to capture, categorize, or remediate such attacks effectively.
Innovation Solution
A threat detection platform that monitors digital activities and builds digital profiles for vendor accounts to analyze email context and content, employing machine learning algorithms to identify anomalies and quantify risk, using a federated database to share information across enterprises, and includes modules for profiling, scoring, and cataloging to detect and prevent external fraud.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional security defenses are used to filter emails, then general email security is maintained, but external fraud from compromised vendor accounts cannot be detected
Solution Approach 1:
The security system is segmented into multiple specialized modules: a vendor profile generator that creates digital profiles of legitimate vendors, a threat detection platform that analyzes incoming emails, and a federated database system that stores and shares fraud information. This segmentation allows each module to specialize in specific aspects of fraud detection, improving reliability without requiring a complete overhaul of the entire email security system.
Solution Approach 2:
The patent introduces an intermediary threat detection platform that sits between the email filtering system and the final recipient. This intermediary layer analyzes emails from compromised vendor accounts by comparing them against digital profiles and fraud indicators, enabling detection of external fraud without replacing the existing email infrastructure.
2Productivity
If manual fraud analysis is performed by fraud departments, then detailed investigation is possible, but the burden and time consumption increase significantly
Solution Approach 1:
The system performs preliminary actions by automatically generating digital profiles of legitimate vendors and their communication patterns before fraud occurs. When emails arrive, the threat detection platform compares them against these pre-established profiles, enabling rapid identification of anomalies without requiring manual investigation of each email.
Solution Approach 2:
The system implements feedback mechanisms where detected fraud cases are automatically added to the federated database as new fraud indicators. This feedback loop continuously improves the system's detection capabilities, allowing it to become increasingly efficient over time while reducing the need for manual fraud department involvement.
3Measurement precision
If vendor-specific information is not examined, then general email processing is maintained, but external fraud attacks go undetected
Solution Approach 1:
The patent applies local quality by creating vendor-specific digital profiles that contain localized information about each vendor's legitimate communication patterns, domains, and behaviors. The threat detection platform then applies these localized profiles to analyze incoming emails from specific vendors, enabling precise detection of fraud while avoiding the need to analyze all emails with the same complex rules.
Solution Approach 2:
The system changes parameters by transitioning from generic email analysis to vendor-specific parameter analysis. The digital profiles contain specific parameters such as legitimate domains, communication patterns, and vendor-specific identifiers that are used to evaluate incoming emails, thereby improving detection precision without requiring examination of all possible email characteristics.
4Reliability
If a federated database is implemented to share fraud information across enterprises, then fraud detection capability is improved, but information security and privacy concerns increase
Solution Approach 1:
The federated database stores copies of fraud indicators and vendor profile information rather than requiring enterprises to share their actual vendor data. Each enterprise maintains its own vendor profiles locally, while the federated database contains aggregated fraud indicators that can be shared safely across organizations, improving detection reliability without exposing sensitive information.
Data Source
AI summary
Techniques for detecting instances of external fraud by monitoring digital activities that are performed with accounts associated with an enterprise are disclosed. In one example, a threat detection platform determines the likelihood that an incoming email is indicative of external fraud based on the context and content of the incoming email. To understand the risk posed by an incoming email, the threat detection platform may seek to determine not only whether the sender normally communicates with the recipient, but also whether the topic is one normally discussed by the sender and recipient. In this way, the threat detection platform can establish whether the incoming email deviates from past emails exchanged between the sender and recipient.