This repository contains three Jupyter notebooks, each implementing a different recommendation system method. To run the notebooks and obtain results, simply execute the code cells from the beginning to the end. The dataset used for this project is the Amazon Fine Food Reviews Dataset, which can be downloaded externally using the following link:
https://www.dropbox.com/s/jm9lb705vk5itk9/Reviews.csv?dl=0
Please note that the local directories in the source code might be different from your own local setup. To properly run the code, you need to set up your directory to access the downloaded dataset from Kaggle.
- CF_final.ipynb: Implements traditional Collaborative Filtering method.
- Food_Review_SVD++.ipynb: Implements Singular Value Decomposition ++ (SVD++) method.
- Neural Collaborative Filtering.ipynb: Implements Neural Collaborative Filtering (NCF) method.
To run the notebooks, you need to have the following software and libraries installed:
- Python 3.x
- Jupyter Notebook
- NumPy
- pandas
- os
- scikit-learn
- Matplotlib
- keras
- Tensorflow
- nltk
- surprise
Any additional libraries specified in the notebooks.
Clone the repository or download the notebooks to your local machine. Open a terminal or command prompt and navigate to the folder containing the notebooks. Run jupyter notebook to start the Jupyter Notebook server. Open the desired notebook in your web browser and follow the instructions within the notebook.
After downloading the Food Reviews dataset mentioned above, place it in the appropriate directory in your local environment. Then, update the file paths in the source code to match your local directory structure. This will ensure that the code can access the dataset correctly when executed.
Execute the code cells in order, from the beginning to the end of the notebook, by clicking "Run" or pressing Shift + Enter.