Project 23 of 39 · Data
Conditions Data Pipeline
Scraper that walks every condition page on ada.com and extracts each section heading with the content beneath it, covering symptoms and causes, then writes the result to JSON and CSV for downstream medical NLP work.
01Overview
A scraper that walks every condition page on ada.com and extracts each section heading together with the content beneath it, covering symptoms, causes and types.
The extraction follows the page's own structure: headings mark the boundaries, and the blocks between them are the content.
02How it is built
- Outputs JSON and CSV, so the result is usable for analysis and for medical language processing.
- Structure is preserved rather than flattened, so a symptom stays attached to its condition.
- Built as source material for retrieval and modelling work.