// the find
Nikhilkohli1/Digital-Marketing-Analytics
This contains projects based on Algorithmic Marketing like Marketing Mix Modeling, Attribution Modeling & Budget Optimization, RFM Analysis, Customer Segmentation, Recommendation Systems, and Social Media Analytics
A grad-school course portfolio (Northeastern INFO7374) packaged as a GitHub repo, walking through marketing mix modeling, multi-touch attribution, RFM/customer segmentation, uplift modeling, and recommendation systems (xDeepFM, Customer2Vec) applied to datasets like Olist and Criteo. Useful as a reference implementation gallery for someone learning algorithmic marketing techniques in Python, not as a library you'd install or extend.
Decent breadth of techniques in one place — adstock/MMM, several attribution model variants (logistic regression, LSTM with attention, position/time-decay heuristics), RFM segmentation, and two different recommendation approaches (ALS/SVD and embedding-based Customer2Vec). Uses a real, messy e-commerce dataset (Olist) rather than a toy one, which makes the notebooks more representative of what these techniques look like on actual data.
Dead since April 2020 with zero commits since — any Heroku demo links or the Tableau Online dashboard referenced in the README are almost certainly down. The README embeds a plaintext username/password for the Tableau dashboard, which is a bad practice to copy even for a class project. It's a flat dump of assignment folders with PDFs, screenshots, and raw CSVs committed to git rather than a structured package — no requirements.txt at the root, no setup instructions, no tests, and no single entry point, so pulling any one technique out means hunting through a specific assignment's notebook and guessing at its dependencies.