Datasetpublic

drdavidprivacy/practicum-case-packs

AI Governance Practicum: Case Pack Extracts Data files used by the Colab notebooks of The AI Governance Practicum (Dr. David, LLC). Each file is presented inside the course as the historical records of a fictional organization. The organizations, systems, and people in the course are fictional. The data is real, public, and reused under its original license. File in this repo Fictional organization and system Source dataset Citation ledgestone_training_extract.csv… See the full description on the dataset page: https://huggingface.co/datasets/drdavidprivacy/practicum-case-packs.

sourceHugging Facecc-by-4.0updated 18h agoView on Hugging Face

1likes70downloads

Dataset Card

AI Governance Practicum: Case Pack Extracts

Data files used by the Colab notebooks of The AI Governance Practicum (Dr. David, LLC). Each file is presented inside the course as the historical records of a fictional organization. The organizations, systems, and people in the course are fictional. The data is real, public, and reused under its original license.

File in this repo	Fictional organization and system	Source dataset	Citation
`ledgestone_training_extract.csv`	Ledgestone Bank, CreditPath	Statlog (German Credit Data), UCI Machine Learning Repository, id 144. Column names and category labels rewritten in plain language from the original documentation; values unchanged.	Hofmann, H. (1994). Statlog (German Credit Data) [Dataset]. UCI Machine Learning Repository. https://archive.ics.uci.edu/dataset/144/statlog+german+credit+data
`corvana_modeling_extract.csv`	Corvana Bank, LimitWise	Default of Credit Card Clients, UCI Machine Learning Repository, id 350. Label column renamed `default_payment_next_month`; values unchanged.	Yeh, I-C. (2016). Default of Credit Card Clients [Dataset]. UCI Machine Learning Repository. https://archive.ics.uci.edu/dataset/350/default+of+credit+card+clients
`alderwyn_encounter_extract.csv`, `alderwyn_code_tables.csv`	Alderwyn Health, CarePivot	Diabetes 130-US Hospitals for Years 1999-2008, UCI Machine Learning Repository, id 296. Unchanged (the code table is the original IDs_mapping file).	Clore, J., Cios, K., DeShazo, J., & Strack, B. (2014). Diabetes 130-US Hospitals for Years 1999-2008 [Dataset]. UCI Machine Learning Repository. https://archive.ics.uci.edu/dataset/296/diabetes+130-us+hospitals+for+years+1999-2008
`crosswind_pilot_extract.csv`	Crosswind Workforce Partnership, ReachFirst	Adult (Census Income), UCI Machine Learning Repository, id 2. Train and test files combined, header row added, trailing periods removed from labels; values unchanged.	Becker, B. & Kohavi, R. (1996). Adult [Dataset]. UCI Machine Learning Repository. https://archive.ics.uci.edu/dataset/2/adult

License

All four source datasets are released by the UCI Machine Learning Repository under the Creative Commons Attribution 4.0 International license (CC BY 4.0). This repository redistributes them under the same license with attribution to the creators and the repository above. The renamed files are derivative works in the sense that column names and labels were rewritten; no data values were altered except as stated in the table.

Use in the course

Learners in the Practicum do not download these files directly. Each week's notebook loads the file it needs by URL and shows the analysis with the code visible. The fictional framing (organization names, system names, and the business context) lives in the course materials, not in the data.

Remediated working files (Week 3 output, Week 4 input)

Added 2026-09-25. Each Week 3 notebook ends by writing the working file after remediation, and these are those files, published so the Week 4 modeling notebook can load them by URL. Each has a README of its own beside it with the row and column arithmetic, the model inputs, the held-out fields, the open items, and the MD5. They hold rule-based changes only (rows removed under a written rule, fields added beside their sources, the outcome column). Nothing in them is fitted, and no value in a column as loaded was altered.

File in this repo	Built from	By	Rows and columns
`corvana_modeling_extract_remediated.csv`	`corvana_modeling_extract.csv`	Corvana LimitWise - Week 3 notebook, section 12	29,995 rows, 40 columns
`alderwyn_encounter_extract_remediated.csv`	`alderwyn_encounter_extract.csv`	Alderwyn CarePivot - Week 3 notebook, section 12	99,340 rows, 58 columns
`crosswind_pilot_extract_remediated.csv`	`crosswind_pilot_extract.csv`	Crosswind ReachFirst - Week 3 notebook, section 12	48,790 rows, 18 columns

The same CC BY 4.0 license and attribution apply. The Ledgestone extract has no remediated file, because that case pack was retired before the first cohort.