Unsupervised machine learning system selection method based on metamorphic testing
A technology of machine learning and metamorphosis testing, applied in the computer field, can solve problems such as poor performance and inability to select clustering algorithms
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0047]The purpose of the present invention is to solve the oracle problem existing in the unsupervised machine learning system (for the clustering algorithm), and the lack of evaluation standards, resulting in the inability to select a suitable clustering algorithm or the technical problem of poor effect, and provides a metamorphosis-based A selection method for unsupervised machine learning systems for testing. The present invention adopts metamorphosis testing technology to alleviate the oracle problem existing in machine learning testing, and through the characteristic analysis of unsupervised machine learning system, proposes 11 general metamorphic relations, and these relations cover the characteristic generally expected by users that machine learning system should have, thus Use these metamorphic relations to verify whether an unsupervised machine learning algorithm meets the requirements; secondly, users define the clustering system evaluation scheme according to their o...
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