The eCommTouch (e-commerce touch) dataset can be used for touch-bassed authenitcation.
This dataset includes touch data from 105 users.
You can download the dataset at Releases.
All data of eCommTouch is stored in the data folder.
The data.zip file contains the following two CSV files:
rawdata.csv: A file that compiles touch data obtained using the iPhone 8 Plus and iPhone XS.user_info.csv: A file that compiles user information.
The features.zip file contains the feature file (features.csv) extracted for the baseline evaluation of eCommTouch.
All programs for feature extraction are located in the src folder.
│── README.md
├── data (After unzipping the 'data.zip' file)
│ │── README.md
│ │── rawdata.csv
│ └── user_info.csv
│
│── features (After unzipping the 'features.zip' file)
│ │── features.csv
│ │ (After executing 'src/datasets_xxx.ipynb')
│ │── 8PLUS_ALL_NORM.csv
│ │── 8PLUS_ALL.csv
│ │── 8PLUS_HOR_NORM.csv
│ │── 8PLUS_HOR.csv
│ │── 8PLUS_VER_NORM.csv
│ │── 8PLUS_VER.csv
│ │── ALL_ALL_NORM.csv
│ │── ALL_ALL.csv
│ │── ALL_HOR_NORM.csv
│ │── ALL_HOR.csv
│ │── ALL_VER_NORM.csv
│ │── ALL_VER.csv
│ │── XS_ALL_NORM.csv
│ │── XS_ALL.csv
│ │── XS_HOR_NORM.csv
│ │── XS_HOR.csv
│ │── XS_VER_NORM.csv
│ └── XS_VER.csv
│
└── src
│── datasets_8PLUS.ipynb
│── datasets_ALL.ipynb
│── datasets_XS.ipynb
└── feature_extraction.ipynb
Python 3.7.8
- pandas==1.1.4
- numpy==1.19.4
- tqdm==4.52.0
- Masashi Kudo
- Hayato Yamana
- Tsubasa Takahashi
The research was supported by NII CRIS Joint Research 2022.