This project builds an AI-driven recommendation system for quick-commerce platforms using collaborative filtering. It analyzes synthetic transactional data from Indian cities to suggest personalized products, demonstrating skills in data analysis, machine learning, and automation.
The system uses a KNN model with cosine similarity to generate recommendations based on purchase patterns.
- Objective: Provide tailored product suggestions to boost user engagement and sales in e-commerce.
- Dataset: Synthetic 1,000-record CSV with varied transactions across 10 Indian cities (e.g., higher in Mumbai).
- Technologies: Python (Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn).
- Business Impact: Can increase conversion rates by 15-25% through personalized recommendations.
-
Clone the repository:
git clone https://github.com/yourusername/ai-recommendation-system-quick-commerce.git cd ai-recommendation-system-quick-commerce -
Install dependencies:
pip install -r requirements.txt
-
Open the notebook:
- Use Jupyter Notebook or Google Colab to run
notebooks/recommendation_system.ipynb.
- Use Jupyter Notebook or Google Colab to run
- Run the Notebook: Execute cells step-by-step to clean the data, build the model, and visualize recommendations.
- Import Data: In the notebook, use
pd.read_csv('Quickcommerce_Data.csv')to load the dataset (no terminal commands needed for import). - Example Output: For 'Grocery Pack', the system might recommend 'Organic Fruits' with a similarity score of 0.54.
- Flowchart: See the end-to-end process below.
- PDF Report: For a detailed walkthrough, including methodology, results, and conclusions, check
project_report.pdf. - Conclusion Summary: This project showcases practical AI application, improving e-commerce efficiency. It highlights skills in data handling and ML, suitable for remote automation roles.
This project is licensed under the MIT License - see the LICENSE file for details.
For questions or collaborations, reach out via LinkedIn