PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 18, 2026Concurrency and Computation Practice and Experience0 citations

Interactive Web API Recommendation Using LLM and Multi-Round Reward Feedback

Interactive Web API Recommendation via LLM‐Enhanced Semantics and Multi‐Round Reward Feedback

View Full Paper
Ask AI
Bookmark
Share

Authors

GKGuosheng KangZCZhuo ChenJLJianxun Liu

Discussion

Loading...

Member takes

Overview

This framework enhances Web API recommendations via dynamic feedback, improving mashup development accuracy.

Key Points

  • The aim is to enhance Web API recommendations by capturing developers' intentions and using dynamic user feedback.
  • Developed the iLLMRec framework incorporating a large language model for semantic enhancement.
  • Created a feature extraction module using GloVe, CNN, and self-attention for deeper semantic correlation analysis.
  • Implemented a multi-round reward feedback mechanism for dynamically optimizing API recommendations.
  • Formulated API recommendations as a multi-label classification problem with Binary Cross-Entropy Loss for parameter adjustment.
  • iLLMRec outperformed state-of-the-art methods in precision, recall, F1-score, and NDCG metrics.
  • The combination of LLM-based enhancement and interactive feedback significantly increased recommendation accuracy.

Cite This Study

Kang et al. (2026) studied this question.

synapsesocial.com/papers/69e3201440886becb653f340https://doi.org/10.1002/cpe.70704
View Full Paper
Ask AI
Bookmark
Share