.png){kind=icon}
Return
User Feedback-Driven Generative Models-Based Methodology for Service Construction
.png){kind=icon}
PRE
AI
DOI：10.1109/MIC.2025.3548143.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Software services are dynamic and can be composed or mashed up to create applications, with user feedback driving continuous optimization. In data-driven intelligent software development, supporting the composition and optimization of these services based on user–developer interaction data is a key challenge. Issues such as inaccurate service recommendations, unstructured feedback, and the ongoing evolution of services complicate this process. This article proposes a solution that leverages software service recommendations to construct mashups, extracts valuable insights from user feedback, and optimizes services in response. Optimization is addressed at both the service and code levels to alleviate developer workload. A case study demonstrates the service construction process, illustrating how these methods improve service composition and performance.
Keywords:
Optimization
Codes
Mashups
Internet
Iterative methods
Data mining
Translation
Web sites
Training
User experience
Generative AI
Software as a service
Feedback
Journal
IF:
4.4
Papers:
2.0K
Citations:
2.0K
Organization
Cited Papers
.png){kind=icon}
ShareSave
ShareSaveDySR: A Dynamic Graph Neural Network Based Service Bundle Recommendation Model for Mashup Creation
PREAI
ShareSaveShareSave
View more