Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2401.08673 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- Crowdsourcing data science competitions has become popular as a cost-effective alternative to solving complex energy-related challenges. How-ever, comprehensive reviews on hosting processes remain scarce. Therefore, this paper undertakes a detailed review of 33 existing data competitions and 12 hosting platforms, complemented by an in-depth case study of the ADRENALIN load disaggregation competition. The review identifies essential elements of competition procedure, including platform selection, timeline, datasets, and submission and evaluation mechanisms. Based on proposed 16 evaluation criteria, the similarities and differences between data competition hosting platforms can be categorized into platform scoring and popularity, platform features, community engagement, open-source platforms, region-specific platforms, platform-specific purposes, and multi-purpose platforms. The case study underscores strategic planning's critical role, particularly platform selection. The case study also shows the importance of defining competition scope which influences the whole com-petition content and procedure, especially the datasets.