Case Study
HR Database Management (BBL)
Lab pipeline that turns HR .docx files into structured CSV and scrapes public contact emails across a three-notebook workflow.
Executive Summary
A lab project for Dr. Danling Jiang's Blockchain Business Lab that turns HR .docx documents into structured CSV databases and enriches them with public contact info via web scraping.
Problem & Constraints
HR documents arrive as unstructured .docx files. The lab needed a reproducible workflow to extract professional profiles (name, affiliation, education, domain) and add contact emails from public sources.
Architecture
Docx_editor.ipynb → HR_database.ipynb → HR scraping.ipynb → structured CSV with email provenance.
Methodology
- Docx_editor: Upload, merge, and clean multiple .docx files
- HR_database: Parse cleaned text into structured fields (name, affiliation, PhD/Masters flags, domain)
- HR scraping: Search Google Scholar, professional websites, and Google queries for public emails
Results & Metrics
| Metric | Result |
|---|---|
| Pipeline stages | 3 notebooks |
| Output | Structured U.S. professional CSV |
| Email tracking | Source column per contact |
| Repository | SBU-BBL/HR-Database (private) |
Tech Stack
Python, Jupyter, python-docx, BeautifulSoup, Pandas, Google Colab
Future Work
Automated scheduling, validation rules for email confidence scoring, integration with lab CRM.