A powerful web-based dashboard that allows users to upload student or individual name/university data and automatically search for potential LinkedIn matches. It uses advanced semantic similarity via MPNet, fuzzy matching, SerpAPI queries, and Selenium-based Bing fallbacks. Results include estimated location and income, when available.

• Upload a CSV of names and universities, or enter data manually.
• Multi-threaded search using SerpAPI and Bing.
• Intelligent matching with SentenceTransformer and RapidFuzz.
• Auto-detection of graduation year, estimated location, and income.
• Confidence scoring using cosine similarity and fuzzy ratios.
• Clean UI with progress tracking and CSV export.

Install the necessary Python packages using:

pip install pandas dash chardet selenium requests sentence-transformers rapidfuzz webdriver-manager numpy

You will also need:

• Chrome installed (for Selenium headless browser)
• SerpAPI key (you can add your free API key in the code or via environment variable)

python your_script_name.py

A Chrome app window will open with the dashboard running at http://127.0.0.1:8050/.

• Free SerpAPI accounts are limited to 100 queries/month. The app will automatically fall back to Bing scraping after that.
• Locations and income are estimated based on keywords and may not always be accurate.
• Manual mode lets you search a single person and enables a "🔎 Manual Mode" indicator.
• CSV format should include columns like: First Name, Last Name, University, Graduation Year (optional).

This project was created entirely by:

• Arik Gitman
• Soham Desai
• Andrew Jerome
• Dylan Shah
• Darsh Patel

• Add GitHub integration and auto-deployment.
• Improve accuracy of location estimation.
• Add LinkedIn scraping if logged-in cookies are provided.
• Use OpenAI embeddings or LLMs for fallback summaries.