API Configuration
Set up via cPanel → Setup Python App. The URL is your domain + app path.
Drop PDF files here
or browse files — up to 20 PDFs
Processing Pipeline
Upload PDFs, extract text, then enrich with Claude AI. Match links renamed PDFs to database entries.
|
Max:
Processing...0/0
Database Maintenance
Diagnose coverage gaps, find duplicates, and clean up your database.
OCR Processing
Some PDFs are scanned images with no extractable text. These need OCR before they can be processed.
How to OCR scanned papers
RECOMMENDED Adobe Acrobat Pro — best quality for academic papers with equations, tables, and multi-column layouts.
- Click Download ZIP for OCR below
- Extract the ZIP on your computer
- In Acrobat: Tools → Action Wizard → Recognize Text on the folder
- Upload the OCR'd PDFs back to the server (replace originals in papers/)
- Run Extract Text → Enrich in the Pipeline above
FALLBACK ocr_papers.py — free, open-source alternative using OCRmyPDF + Tesseract.
pip install ocrmypdf
# Also install Tesseract: brew install tesseract (mac) / apt install tesseract-ocr (linux)
python ocr_papers.py /path/to/extracted_zip/
PDF Operations
Processing...0/0
Extracted Text Management
Manage extracted text files: match to JSON entries, rename to canonical format, find and remove duplicates.
Diagnostics
Cleanup
Processing...0/0