Software Quality Determination Using Bug Detection Rate and Test Viewpoint Coverage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for determining software bug convergence require previous bug occurrence data, limiting their applicability, and may inaccurately indicate convergence, leading to unexpected contingencies due to potential untested viewpoints.
Innovation Solution
A software quality determination apparatus and method that calculates bug detection rates for each test viewpoint and determines convergence based on whether the rate is equal to or smaller than a reference value after reaching a specified execution amount, without relying on previous data, thereby preventing accuracy reductions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If bug convergence determination is performed using previous bug occurrence result data, then the determination can be made based on historical trends, but the technique can only be applied to systems that have determined bug convergence before
Solution Approach 1:
The patent segments the bug convergence determination process into two independent approaches: one using historical bug occurrence data and another using current test viewpoint coverage. This segmentation allows the system to select the appropriate method based on data availability, resolving the contradiction between reliability and adaptability.
Solution Approach 2:
The patent performs preliminary classification of test viewpoints into executed and non-executed categories before making convergence determination. This preliminary action enables the system to identify untested viewpoints that might harbor undetected bugs, improving determination accuracy without requiring historical data.
2Adaptability or versatility
If bug convergence is determined when bug detection rate increase approaches zero, then the determination can be made without previous bug occurrence data, but there is a probability that untested viewpoints are not being tested
Solution Approach 1:
The patent introduces feedback by continuously monitoring both bug detection rates and test viewpoint coverage. The convergence determination is adjusted based on feedback from untested viewpoints, ensuring that convergence is not declared prematurely when critical viewpoints remain unexplored.
Solution Approach 2:
The patent adds a new dimension to bug convergence determination by incorporating test viewpoint coverage analysis alongside traditional bug detection rate metrics. This dimensional expansion allows the system to assess convergence from multiple perspectives, improving reliability without sacrificing adaptability.
3Reliability
If additional test execution is performed after bug convergence determination, then undetected bugs from untested viewpoints may be found, but this increases test execution time and resources
Solution Approach 1:
The patent performs preliminary identification of untested viewpoints before convergence determination. By knowing in advance which viewpoints remain untested, the system can make informed decisions about whether additional testing is necessary, balancing detection completeness with time efficiency.
Solution Approach 2:
The patent applies partial action by performing convergence determination based on executed viewpoints first, then selectively adding tests for critical untested viewpoints only if convergence is questionable. This avoids exhaustive re-testing while maintaining sufficient detection completeness.
Data Source
AI summary
A software quality determination apparatus includes: a bug detection rate calculator configured to calculate, for each test viewpoint, which is a viewpoint when a system is tested, a detection rate of a bug generated in a test of the test viewpoint; and a bug convergence determination unit configured to determine, for the each test viewpoint, the bug convergence of the test viewpoint depending on whether the calculated detection rate of the test viewpoint is equal to or smaller than a reference value, which serves as a reference of the detection rate of the test viewpoint after a test execution amount of the test viewpoint has reached an execution reference amount, which serves as a reference of the execution amount of the test viewpoint.