Designing Accessible and Rigorous ML Experiments: No-Code Visual Workflows and Structured Guidance for Beginners
4.00T1 sourcearXiv cs.HC
Source record
Published by arXiv cs.HC (T1 source). The original is at https://arxiv.org/abs/2611.13386.
Pipeline notes
The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.
SummaryTwo within-participant studies with 18 predominantly technically trained participants compared mlguide’s structured workflow guidance and AI Sandbox’s visual no-code interface with notebook-based alternatives. Guidance improved high-level workflow designs, while no-code use improved hedonic ratings but not workload or acceptance; methodological-quiz scores did not significantly improve.
Why it mattersThe paper separates workflow-planning support from execution-environment support and supplies reusable assessment materials for adapting the approach.
Cited by
No citations on record.