hashan-7/medscript-drug-dataset

---
license: mit
language:
- en
tags:
- healthcare
- nlp
- ocr
size_categories:
- 10K<n<100K
---
# MedScript Drug Dataset

## Description
This dataset contains cleaned and consolidated medicine names collected from multiple sources. It is designed for use in prescription OCR post-processing and drug name matching.

## Structure
- drug_name: normalized medicine name (lowercase, cleaned)

## Sources
- product dataset
- drug_database.csv
- massive_drug_list.csv
- sl_drugs dataset

## Processing
- converted to lowercase
- trimmed whitespace
- removed duplicates
- removed invalid entries

## Use Case
- OCR correction
- fuzzy matching
- medical text normalization