Neural Network Classifier for Automated Software Defect Assignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing volume of software defect reports in modern software products leads to a significant delay in assigning developers to resolve defects, as the manual triage process is time-consuming and inefficient.
Innovation Solution
A feedforward neural network classifier is trained to quickly and accurately assign developers to software defect reports by extracting features from summary, description, and comments fields, and transforming them into a vector space model for automated classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual triage process is used to assign developers to software defect reports, then assignment accuracy can be maintained through human judgment, but the processing time increases significantly and productivity decreases
Solution Approach 1:
The patent replaces the manual mechanical triage process with an automated machine learning system. A feedforward neural network classifier processes defect report features and automatically assigns developers, eliminating the need for manual human review while maintaining or improving assignment accuracy through learned patterns from historical data.
Solution Approach 2:
The patent introduces a machine learning model as an intermediary between defect report creation and developer assignment. This intermediary system processes defect features, extracts relevant information, and generates assignment recommendations, serving as a bridge that automates the triage process while preserving the ability to achieve accurate assignments.
2Reliability
If manual triage process is used to review and assign developers to software defect reports, then quality control can be maintained, but the time from defect discovery to assignment increases from days to weeks
Solution Approach 1:
The patent performs preliminary actions by pre-training the machine learning model on historical defect data and developer performance information. The model is prepared in advance to quickly process new defect reports, enabling rapid assignments without compromising quality. The system extracts features and generates recommendations automatically as soon as a defect is created.
Solution Approach 2:
The patent substitutes the slow manual review process with an automated machine learning system that processes defect reports in minutes rather than days. The neural network classifier rapidly evaluates defect features and recommends appropriate developers, dramatically reducing the time from defect discovery to assignment while maintaining reliable quality through learned patterns.
3Productivity
If automated machine learning classification is implemented for developer assignment, then processing speed and productivity improve, but system complexity increases due to model training and feature extraction requirements
Solution Approach 1:
The patent segments the developer assignment system into distinct modular components: feature extraction module, model training module, and classification recommendation module. This segmentation allows each component to be developed, maintained, and optimized independently, reducing overall system complexity while enabling high processing speed through specialized functionality in each module.
Solution Approach 2:
The patent creates a universal machine learning framework that can handle multiple defect types and assignment scenarios through a single trained model. The system extracts general features from diverse defect reports and applies the same classification logic across different situations, reducing complexity by avoiding the need for separate specialized systems for each defect category.
4Measurement precision
If features are extracted from multiple fields including summary, description, and comments, then assignment accuracy improves through comprehensive analysis, but processing complexity and time increase
Solution Approach 1:
The patent extracts only the most relevant features from defect report fields such as summary, description, and comments. The machine learning model identifies and extracts key information elements that are most predictive of appropriate developer assignment, ignoring redundant or less important information. This selective extraction maintains high assignment accuracy while reducing processing complexity compared to analyzing all possible fields.
Data Source
AI summary
A technique includes processing, by a computer, data representing a software defect report to extract features from the software defect report. The software defect report contains information that identifies a defect in a software product. The technique includes applying, by the computer, a feedforward neural network classifier to the features to identify a developer to assign to the identified defect.