Data Science Consulting: PDF Parser

PDF Query extracts information from 9,000+ CommonApp student applications, each spanning 15+ pages, totaling 213,000,000+ words.

Raw text extraction from each PDF page using PyPDF2.

78+ Regex patterns are written to generate specific matches, feeding key fields into a structured data frame.

Regex-based field extraction for citizenship and visa status, with fallback handling.

Pandas assembles the final 101-column data frame. Due to privacy concerns, the final sheet can't be shown; trust is implored instead.

Assembling each applicant's extracted fields into a structured output row.

Pandas converts the final data frame to Excel for delivery: 9,000+ PDFs become 9,000+ rows.