This project explores the power of feature engineering in the context of opportune moment detection for interrupting smartphone users. It introduces a novel feature engineering approach named Contextual Filtered Features (CFF) and includes a robust machine learning setting for proper evaluation, such as using Leave-One-Subject-Out.
- Clone the repository and navigate to the project folder:
git clone [email protected]:Jumabek/receptivity.git
cd receptivity
- Link your local Kemophone data folder:
ln -s /home/juma/dataonssd/kemophone data/
Explore the Jupyter notebooks located in the analysis
folder for insights into the data analysis and feature engineering process.
[Description of the data structure, content, and any important details.]
If you use our data or code for your research, please cite our paper:
@article{alikhanov2023design,
title={Design of Contextual Filtered Features for Better Smartphone-User Receptivity Prediction},
author={Alikhanov, Jumabek and Zhang, Panyu and Noh, Youngtae and Kim, Hakil},
journal={IEEE Internet of Things Journal},
year={2023},
publisher={IEEE}
}
This project is open-sourced under a free license.
For questions or collaboration opportunities, contact [email protected].
This research outcome would not be possible without the support and mentoring of Prof. Uichin Lee from KAIST.