Cross-project defect prediction method based on feature distribution alignment and neighborhood instance selection
A feature distribution and prediction method technology, applied in character and pattern recognition, software testing/debugging, error detection/correction, etc., to improve defect prediction performance and reduce distribution differences
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment
[0035] The present invention comprises the following steps in practical application:
[0036] Step 1: Select source items from the software defect dataset, merge all source items to form source item set D S , and select a target item D from the software defect data T ;where target item D T and the selected source item set D S The same module data does not exist, that is, it is not a dataset of different versions of the same project;
[0037] Step 2: Calculate the source itemset D S The covariance matrix C of S , calculate the target item D T The covariance matrix C of T , and the corresponding calculation formula is:
[0038] C S =COV(D S )
[0039] C T =COV(D T )
[0040] Step 3: As figure 2 As shown, perform feature distribution alignment, specifically by first eliminating the source itemset D S The correlation between features is calculated as:
[0041]
[0042] Then put the target item D T The feature correlations are populated to eliminate source item...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com