Datasets:
metadata
license: cc-by-4.0
task_categories:
- tabular-classification
- tabular-regression
language:
- en
tags:
- healthcare
- medicine-quality
- online-pharmacy
- e-pharmacy
- social-media
- darknet
- falsified
- consumer-safety
- sub-saharan-africa
- lmic
pretty_name: >-
Online Pharmacy & E-Pharmacy Regulation (SF Prevalence, Platform Type,
Consumer Safety)
size_categories:
- 10K<n<100K
configs:
- config_name: licensed_e_pharmacy
data_files: data/epharmacy_licensed_e_pharmacy.csv
- config_name: social_media_marketplace
data_files: data/epharmacy_social_media_marketplace.csv
default: true
- config_name: rogue_website_darknet
data_files: data/epharmacy_rogue_website_darknet.csv
Online Pharmacy & E-Pharmacy Regulation Dataset
Abstract
30,000 simulated online medicine purchase observations (10,000 per scenario) across three e-pharmacy platform types. Variables include platform, product category, seller licensing, prescription verification, SF classification, price, payment method, and consumer demographics. Three scenarios: licensed e-pharmacy (6% SF), social media marketplace (65%), rogue website/darknet (89%).
This dataset is entirely simulated. It must not be used for regulatory or enforcement decisions.
Validation
Usage
from datasets import load_dataset
dataset = load_dataset("electricsheepafrica/online-pharmacy-regulation", "social_media_marketplace")
df = dataset["train"].to_pandas()
print(df.groupby('platform_name')['quality_test_result'].apply(lambda x: (x=='fail').mean()))
References
- WHO (2024). SF medicines sold online.
- NABP (2023). 95% of online pharmacies non-compliant.
- Interpol Operation Pangea. Illicit online pharmacy crackdowns.
- LegitScript. 30,000+ rogue pharmacy websites.
Citation
@dataset{esa_epharmacy_2025,
title = {Online Pharmacy and E-Pharmacy Regulation Dataset},
author = {{Electric Sheep Africa}},
year = {2025},
publisher = {Hugging Face},
url = {https://huggingface.co/datasets/electricsheepafrica/online-pharmacy-regulation}
}