This repository contains the implementation of the model presented in the paper "Modelling Concept Drift in Dynamic Data Streams for Recommender Systems".

The repository includes the following files:

• DataStream_AMZ_exp.ipynb: Executes experiments on the Amazon Video Games dataset.
• DataStream_AMZ_INV_exp.ipynb: Executes experiments on the Amazon Video Games dataset with inverted roles.
• DataStream_MIND_exp.ipynb: Executes experiments on the MIND dataset.
• DataStream_YOOCHOOSE_exp.ipynb: Executes experiments on the Yoochoose dataset.
• datastream.py: Contains the common code used across all experiments.
• Readme.md: this files.
• Dockerfile: docker file to build a custom image with a JupyterLab instance.
• requirements.txt: dependencies file.
• MSNewsDatasetPreprocess.ipynb: notebook for the MIND dataset preprocessing.
• PreprocessYoochooseDataset.ipynb: notebook for the Yoochoose dataset preprocessing.

To run the experiments, follow these steps:

1. Clone the Repository:

   git clone https://github.com/fsp22/mcd_dds4rs.git
   cd mcd_dds4rs
   # fix permission for docker environment
   chmod 777 your-repo-name
   chmod 666 your-repo-name/*

2. Build the docker image (optional)

   docker build --rm --tag my-ds-image .
   

   The docker contains required libraries for this project. The dependecies are defined in the requirements.txt.

3. Executes the Jupyter notebooks.

   docker run -it --rm -p 18888:8888 -v "${PWD}":/home/jovyan/work my-ds-image start-notebook.py --ip 0.0.0.0 --IdentityProvider.token=''

4. Open the browser at

   http://localhost:18888/lab
   

Now you can execute the notebooks.

For the dataset Yoochoose and MIND, the source dataset files must be pre-processed with the notebooks:

• MSNewsDatasetPreprocess.ipynb

• PreprocessYoochoseDataset.ipynb