AI Datasets can serve multiple phase purpose:
-1) training dataset
To learn the weights of the neurons
-2) checking dataset
An independent Dataset assessing that the trained model works properly for the problem studied
-3) operational real life dataset
Once the model is trained and checked, using it in automatic/independent mode.
Are there criteria on such sets to ensure they are fit for purpose?
Size, quality, homogeneity, représentativity, statistical relevance?
Dear Renaud Di Francesco'
In the conclusion to our year-end series, Synced spotlights ten datasets that were open sourced in 2018 and takes a peek into the papers behind them. We hope this list can provide the AI community with insight into what 2019 might hold in store for big data.
https://medium.com/syncedreview/2018-in-review-10-open-sourced-ai-datasets-696b3b49801f
Data Web Of Science Dataset
Regards,
Shafagat
Thanks Shafagat Mahmudova for the inventory of some data sets made available.
This reinforces the case for a common methodology, criteria, formats, UI for the datasets types mentioned in the question, indeed!
- Badges
- Science topic
- Similar topics
- Organizational Psychology
- Training