Graphical Neural Network for Enterprise System Error Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The process of detecting and correcting errors introduced during application upgrades in computing systems is inefficient and time-intensive, often resulting in undetected and uncorrected errors that negatively impact the application and the system.
Innovation Solution
A computing platform uses a graphical neural network to analyze system and virtual parameters, dynamically link nodes to identify discrepancies, and output actions to an AI engine for error correction, updating virtual nodes and executing commands to correct errors in a simulated environment before applying changes to the actual system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual review methods are used to detect errors after application upgrade, then detection accuracy can be maintained, but the process becomes inefficient and time-intensive
Solution Approach 1:
The patent creates a simulated enterprise system before the actual application upgrade to pre-detect potential errors. By performing error detection in advance in the simulated environment, the system identifies issues before they affect the production system, thereby maintaining detection accuracy while reducing the time needed for post-upgrade manual reviews.
Solution Approach 2:
The patent creates a copy of the enterprise system (simulated enterprise system) that mirrors the production environment. This copy is used to test the upgraded application and detect errors without impacting the actual system. The graphical neural network analyzes parameters in this copied system to identify errors, enabling efficient detection without compromising accuracy.
2Measurement precision
If comprehensive error detection is performed on all system parameters, then detection accuracy improves, but system complexity increases
Solution Approach 1:
The patent introduces a graphical neural network as an intermediary between the simulated enterprise system and the error detection process. This neural network automatically analyzes system parameters, virtual parameters, and their relationships to identify errors, comprehensively examining all parameters without requiring complex manual analysis procedures, thus maintaining comprehensiveness while managing system complexity.
Solution Approach 2:
The graphical neural network serves multiple functions simultaneously: it creates the simulated enterprise system, generates system and virtual parameters, establishes relationships between parameters, and performs error detection. This multi-functional approach enables comprehensive error detection across all parameters without proportionally increasing system complexity, as the same component handles multiple tasks.
Data Source
AI summary
Aspects of the disclosure relate to upgrading an application within a simulated version of an enterprise system to detect and correct potential errors as a result of the upgrade. A computing platform may create a simulated version of the enterprise system by receiving metadata associated with the enterprise system, and converting the metadata into system parameters. Virtual parameters may be created by the computing system based on upgrading an application within the simulated version of the enterprise system. The computing system may create system nodes and virtual nodes. The system nodes and virtual nodes may be dynamically linked in order to determine errors caused by the application upgrade within the simulated version of the enterprise system. The computing platform may determine actions to correct the errors and input the results and feedback into an AI engine to further refine the accuracy and reliability of the computing platform over time.