Semantic Graph Visualization for Policy Violation Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Detecting fraud and mistakes in organizational transactions is complex due to the use of multiple computing systems with different data representation and storage methods, requiring intimate knowledge of each system and their relationships, and as perpetrators adapt to new technologies, effective detection becomes difficult.
Innovation Solution
A computer-implemented method that identifies policy violations by analyzing data to find semantic objects related to the violation, displaying graphical representations of these objects in a directed graph, indicating relationships between them, and allowing users to view workflows and attributes, enabling a forensic investigation of policy violations across different systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple computing systems are used for different organizational purposes, then functional versatility is improved, but system complexity and difficulty of detecting policy violations worsen
Solution Approach 1:
The patent introduces a semantic mapping layer that acts as an intermediary between multiple computing systems and the fraud detection system. This layer translates data from different systems into a unified semantic model, allowing the organization to maintain multiple specialized systems while simplifying the detection process through a common interface.
Solution Approach 2:
The patent creates a universal semantic model that can represent data from multiple different computing systems (CRM, HR, accounting, etc.) in a unified framework. This semantic model serves multiple functions: it enables fraud detection, supports forensic investigation, and provides a common language across diverse systems.
2Productivity
If data is stored in optimized formats for efficient access, then data processing speed is improved, but intuitiveness and ease of forensic investigation worsen
Solution Approach 1:
The patent creates a semantic copy or representation of the original data stored in optimized formats. This semantic model preserves the essential relationships and meanings of the data while presenting it in an intuitive graphical format suitable for forensic investigation, without requiring changes to the underlying optimized storage.
Solution Approach 2:
The patent transforms data from the traditional tabular/dimensional storage format into a graphical visualization dimension. By representing semantic objects and their relationships as visual elements in a graphical interface, the system maintains efficient data processing while dramatically improving intuitiveness and ease of investigation.
3Measurement precision
If forensic investigation capabilities are enhanced, then fraud detection accuracy is improved, but the time and resources required for investigation increase
Solution Approach 1:
The patent performs preliminary semantic modeling and relationship mapping during normal system operation, before forensic investigation is needed. By pre-establishing the semantic model and relationships between data elements, the system reduces the time required for actual investigations while maintaining high detection accuracy.
Solution Approach 2:
The patent replaces manual forensic investigation processes with an automated semantic analysis system. The graphical interface and automated relationship tracking substitute for time-consuming manual data correlation efforts, significantly reducing investigation time while improving accuracy through consistent application of semantic rules.
Data Source
AI summary
Techniques for displaying information. Policy violations are identified, based at least in part on data stored in a data store. For the policy violations, a plurality of semantic objects related to the violations are identified. Arrangements of graphical objects are displayed where the graphical objects represent the identified semantic objects and where the arrangement indicates one or more relationships between pairs of the semantic objects.