Privacy-Preserving Personalized Fitness Recommender System (P³FitRec): A Multi-level Deep Learning Approach

This repository contains code, model, dataset for P³FitRec at ACM TKDD.

Guide

Directories	Description
data scalers	Trained data scaler objects
datasets	Datasets used in preprocessing and training (provided as google drive links)
entity embedding	Script for entity embedding training and trained entity embeddings
preprocessing	Scripts for data preprocessing
workout Speed&HeartRate model	Script for speed & Heart rate prediction and trained models
workout distance prediction model	Script for distance prediction training, inference and trained model

Citation

The data, models and its implementation are available for potential improvements and reproducibility under the license as described under ACM. Please make sure to cite the following research publication should you need to use any of the materials provided in this repository.

ACM citation Xiao Liu, Bonan Gao, Basem Suleiman, Han You, Zisu Ma, Yu Liu, and Ali Anaissi. 2023. Privacy-Preserving Personalized Fitness Recommender System (P3FitRec): A Multi-level Deep Learning Approach. ACM Trans. Knowl. Discov. Data Just Accepted (January 2023). https://doi.org/10.1145/3572899

@article{10.1145/3572899,
  author={Liu, Xiao and Gao, Bonan and Suleiman, Basem and You, Han and Ma, Zisu and Liu, Yu and Anaissi, Ali},
  title={Privacy-Preserving Personalized Fitness Recommender System (P3FitRec): A Multi-Level Deep Learning Approach},
  year={2023},
  publisher={Association for Computing Machinery},
  address={New York, NY, USA},
  issn={1556-4681},
  url = {https://doi.org/10.1145/3572899},
  doi = {10.1145/3572899},
  journal={ACM Trans. Knowl. Discov. Data},
  month = {jan},
  keywords = {Personalization; fitness; recommender system; deep learning; sensors}
}