Diagnostic Message Prioritization via ML Exception Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
During software development, non-error diagnostic messages indicating potential defects in source code are often ignored, leading to exceptions during executable file execution, as developers are unaware of their impact and the large number of these messages makes resolution time-consuming.
Innovation Solution
A determination model is trained to identify potentially-problematic diagnostic messages and combinations that may cause exceptions, prioritizing them for resolution, using correlation sets of diagnostic messages and exception messages to predict execution issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all non-error diagnostic messages are resolved, then exception-free execution is achieved, but development time and effort increase significantly
Solution Approach 1:
The system uses machine learning models trained on historical diagnostic message data to predict which messages are likely to cause exceptions. This feedback mechanism allows developers to focus on high-risk messages while automatically filtering low-risk ones, resolving the contradiction between achieving exception-free execution and minimizing development time.
Solution Approach 2:
The system automatically prioritizes and filters diagnostic messages using AI models, reducing the manual effort required by developers. The automated classification service handles the analysis and prioritization, allowing developers to benefit from exception-free execution without proportionally increasing their time investment.
2Measurement precision
If all diagnostic messages are analyzed manually, then accurate identification of problematic messages is achieved, but the process becomes too time-consuming
Solution Approach 1:
The system replaces manual mechanical analysis of diagnostic messages with automated machine learning models. These models process messages at machine speed while maintaining high accuracy through training on historical data, thus substituting the slow manual process with a fast automated system that preserves identification precision.
Solution Approach 2:
The system creates a simplified representation or priority score for each diagnostic message based on patterns learned from historical data. Instead of requiring full manual analysis of each message, the system uses these copied priority indicators to guide developer attention, maintaining accuracy while dramatically improving analysis speed.
3Reliability
If developers focus on resolving all diagnostic messages, then code quality improves, but the large number of messages makes prioritization difficult
Solution Approach 1:
The system segments the large set of diagnostic messages into priority groups based on their likelihood to cause exceptions. By dividing the messages into high, medium, and low priority categories, the system reduces the complexity of prioritization while ensuring that code quality improvements focus on the most critical messages first.
Solution Approach 2:
The system changes the parameter used for message prioritization from manual assessment to an automated priority score derived from machine learning models. This parameter change simplifies the prioritization process by providing an objective, data-driven metric that automatically ranks messages by their potential impact, reducing the complexity of decision-making.
Data Source
AI summary
Example techniques for identification of diagnostic messages corresponding to exceptions are described. A determination model may determine whether a set of diagnostic messages generated based on analysis of a source code includes a diagnostic message that likely corresponds to an exception. The determination may be used to identify a set of diagnostic messages including the diagnostic message that likely corresponds to an exception.