ML Anomaly Detection for Code Architectural Flaws
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional program code testing methods are inadequate for identifying architectural flaws and defects in complex code structures, particularly in large-scale code development where disparate actions by multiple developers can lead to cascading issues, and they are not effective in real-time defect detection and remediation.
Innovation Solution
A machine learning-based anomaly detection system that processes application logs to identify architectural flaws by constructing an application layer map, determining architecture patterns, and detecting anti-patterns, allowing for real-time defect analysis and correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional program code testing methods are used, then the testing process is simple and straightforward, but the ability to identify architectural flaws and defects in complex code structures is inadequate
Solution Approach 1:
The patent introduces an intermediary machine learning model that acts as a mediator between the codebase and the testing process. This model analyzes code metrics, application logs, and architectural patterns to detect defects, thereby enhancing detection capability without requiring the testing system itself to become overly complex. The intermediary model processes the complexity internally while presenting simplified results to users.
Solution Approach 2:
The patent replaces traditional mechanical testing methods with machine learning-based automated analysis. Instead of manual code review or conventional testing mechanisms, the system uses ML models to automatically detect architectural flaws and defects, significantly improving detection precision while managing system complexity through automation.
2Productivity
If manual code review and conventional testing are used, then the system is easy to operate, but real-time defect detection and remediation are not effective
Solution Approach 1:
The patent implements preliminary action by continuously analyzing code changes and application logs in real-time before defects manifest as critical failures. The machine learning model proactively detects architectural flaws and potential defects as they emerge during development, enabling early remediation. This preliminary detection capability increases productivity by preventing defects from propagating through the codebase.
Solution Approach 2:
The system incorporates feedback mechanisms where the machine learning model continuously learns from application logs and code metrics, improving its defect detection accuracy over time. The feedback loop enables the system to adapt to new code patterns and defect types, maintaining ease of operation while enhancing real-time detection capabilities through automated learning and adjustment.
3Productivity
If multiple developers work on the same code simultaneously, then code development productivity increases, but architectural flaws and cascading issues may occur
Solution Approach 1:
The patent introduces an intermediary analysis system that monitors code changes from multiple developers in real-time. This intermediary ML model analyzes the interactions between different code modifications, detects potential architectural conflicts before they become flaws, and provides recommendations to maintain architectural integrity. This enables parallel development productivity while preventing cascading issues.
Solution Approach 2:
The system replaces manual architectural review processes with automated machine learning analysis that can handle multiple concurrent code changes. The ML model automatically detects architectural violations and potential cascading issues arising from simultaneous development activities, maintaining reliability without reducing development speed or requiring manual intervention.
Data Source
AI summary
Embodiments of the invention are directed to machine learning-based anomaly detection in program code. The system provides a machine learning (ML) anomaly detection model component structured to detect architectural flaws in program code based on processing application logs associated with technology program code and determining flow sequences between a plurality of layers of code. In particular, the system trains the machine learning (ML) anomaly detection model that is structured to (i) construct a first application layer map based on mapping each of the plurality of first classes associated with the first technology program code to one or more application layers, (ii) determine a first architecture pattern associated with the first technology program code, and (iii) determine whether the first technology program code is associated with an anti-pattern.