This mini machine learning project was developed as part of a university assignment.
It focuses on discovering association rules in a market basket dataset using three algorithms:
- Apriori
- Eclat
- FP-Growth
After comparing their performance, the FP-Growth algorithm was selected and deployed through a Streamlit web app for interactive rule exploration.
To identify frequent product combinations and generate strong association rules from customer transaction data, helping understand shopping patterns.
| Algorithm | Description | Result |
|---|---|---|
| Apriori | Traditional algorithm for frequent itemset mining | Accurate but slower |
| Eclat | Uses intersection-based approach | Faster than Apriori |
| FP-Growth | Builds a compact FP-tree for mining | ⚡ Best performance (used for deployment) |
- Python
- Pandas / NumPy
- mlxtend (for Apriori, Eclat, FP-Growth)
- Streamlit
- Scikit-learn
- Matplotlib / Seaborn (for visualization)
- Clone the repository
git clone https://github.com/YOUR_USERNAME/ML_mini_projet.git cd ML_mini_projet - use virtual environement
python -m venv venv- install the required dependencies using
pip install -r requirements.txt
pip install streamlit
streamlit run home.py📊 you can find dataSet here : https://www.kaggle.com/datasets/heeraldedhia/groceries-dataset
🌐 you can test the model from here : https://mlminiprojet.streamlit.app/