From Keyword Chaos to Clarity: Conquering a Complex Systematic Review with Python
summary
Automates a scoping review workflow that starts from 25,000 academic abstracts and uses Python regular expressions plus NLP and the Gemini API to extract structured fields, deduplicate and filter; custom scripts convert varied database exports, run 100+ regexes, and drop the pile to 1,611 records. Empirical results show about one third true positives and a 0.4% false‑negative rate, so the system materially reduces manual screening but needs human checks.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 0:59Background
- 4:59Scoping Review
- 5:35Database Search
- 6:41Common Terminology
- 7:50Animal Studies
- 9:35Finding the right outcome
- 10:49Finding relevant articles
- 14:24Finding a delivery method
- 18:45Using the Gemini API
- 26:53Questions
related talks
26mReview Analysis: an Approach to Leveraging User-Generated...
22mPython for science, side projects and stuff!
50mEffective test scrubbing
46mProcessing Hotel Reviews with Python
26mAccelerating COVID-19 insights for clinical use with intelligent search
38mData Science from the Trenches
51mSimplicity For Scale- Analyzing 15 Million DNA Samples With Python
26mAccelerating COVID-19 insights for clinical use with intelligent search