MurtazaMajid/Campbells-AI-and-Marketing-Hub
End-to-end AI marketing intelligence platform: predicts customer churn at 84% AUC, segments 1,500+ customers via RFM KMeans clustering, runs ABSA sentiment analysis, and auto-generates personalised SMS / email / push notifications using LLaMA 3.3-70B via Groq.
SUMMARY AI summary by gpt-5-mini
An end-to-end AI marketing system built for a real restaurant (Campbell’s) that turns transactions, reviews and menu data into actionable customer intelligence. It performs RFM-based customer segmentation (KMeans), a two‑tier churn risk model (XGBoost + rules), aspect-based sentiment analysis (TF‑IDF + logistic regression), behavioral profiling (7 features) and generates personalized re‑engagement messages via an LLM (Groq LLaMA 3.3). Who uses it: restaurant operators, marketing analysts, and data engineers who need to identify at‑risk customers, understand sentiments, prioritize outreach, and automate tailored messaging. Key features: live deployed stack (FastAPI backend, Supabase Postgres, React frontend on Vercel, Railway hosting), interactive dashboard and API endpoints for segmentation/churn/sentiment/customer profiles, pickled ML pipeline, and a dataset of 12,545 transactions across 2,041 customers with labeled reviews and menu data. Tech highlights: Python, FastAPI, React, XGBoost, scikit‑learn.
DETECTED Detected AI stack
AI-related keywords found in this repo's description, topics, or README summary — grouped by category. Each badge links to the corresponding ranking detail page.
Language breakdown (by bytes)
Owner
Final-year Data Science student. Building deployed end-to-end ML systems using NLP, computer vision, and time-series. Currently open to Data / ML / AI roles.
Dates
| Created on GitHub | 2026-04-23 |
| Last push | 2026-05-09 |
| First seen here | 2026-05-09 |
| Last fetched | 2026-05-09 15:42 |
Similar repos (same language)
Beginner-friendly Machine Learning projects built with Python, Scikit-learn and Google Colab.
ak-junior3339/PriceMyRideAI aparna172004/AI-Movie-Recommendation-SystemAn AI-powered Movie Recommendation System built using Machine Learning and Streamlit that suggests movies based on user preferences. The system uses content-based filtering to find similar movies by analyzing movie features such as genres, keywords, cast, and overview.
rmnouv/learning-agentic-ai shady528/StoxAI-powered stock market chatbot with multi-LLM support (Gemini, Claude, GPT). Built with RAG, FastAPI, React, and Vector Databases