Reuse and Automated Integration of Recommenders for Modelling Languages
summary
An Eclipse plugin called IronMan provides a recommendation service API with two endpoints (features and recommend), JSON encodings for targets and items, and automatic mappings (eClass → Class, attributes → properties) so that two existing recommenders get reused across four languages and deploy into six modelling tools without touching editor source. Aggregation supports simple methods like Borda-like ranking, and current integrations target Sirius and tree-based editors with proactive mapping and deployment.
AI-written summary · about
chapters (10)
- 0:00Recommender systems in modeling
- 1:51Recommender integration challenges
- 3:38Reusable recommenders challenges
- 6:05Approach with iroman
- 9:58Adapting recommenders to notations
- 11:20Architecture and tool support
- 13:06Integration demo
- 16:17Evaluation results
- 18:46Conclusion and future work
- 20:32Q&A
related talks
14mAutomating the Synthesis of Recommender Systems for Modelling Languages
15mAutomated Variability Injection for Graphical Modelling Languages
45mAI Assistance for Software Teams: The State of Play
08mHow to Recommend Tools for Finding and Fixing Software Errors
42mIntegrating Developer Experiences
41mAI Tooling for Scala Developers - What JetBrains Brings to the Table
29mYour Plugin for IntelliJ IDEA
52mResistance Is Futile