Dataset icon

Dataset

Common evaluation pitfalls in touch-based authentication systems

Documentation:
This dataset contains the data collected and used in the paper “Common Evaluation Pitfalls in Touch-Based Authentication Systems”. The dataset consists of touchscreen interactions and accelerometer + gyroscope data gathered from iPhone devices. The data was collected remotely through Amazon Mechanical Turk and participants were required to do multiple repetitions of two tasks every day for up to 31 days. In the paper, we use this data to outline evaluation pitfalls in touch-based authentication systems. In total, we collected data from 470 users amounting to 6,017 unique sessions and 1,166,092 unique swipes. All participants included gave consent for sharing their data. The “table.zip” folder contains .csv files with information about the participants such as age, gender, location, etc. Furthermore, it provides details about when the experiments were performed and their resulting filenames. The “data_files.zip” folder contains all the interaction, accelerometer, and gyroscope data in an organized folder structure. Top-level folders represent unique user IDs and contain a list of folders for each daily session. Each daily session folder has a “scroll” (social media task) and “swipe” (image gallery task) folders which contain each repetition done by the user. Each repetition folder, then, hosts the accelerometer, gyroscope, and touch data. The “features.zip” folder contains a single .csv file with extracted features from the data in “data_files.zip” and can be immediately used for touch-based authentication experiments.

Actions

Access Document

Files:

Authors/Creators

More by this author/creator
Institution:
University of Oxford
Division:
MPLS
Department:
Computer Science
Role:
Creator
ORCID:
0000-0002-5558-6497
More by this author/creator
Institution:
University of Oxford
Division:
MPLS
Department:
Computer Science
Role:
Creator
More by this author/creator
Institution:
University of Oxford
Division:
MPLS
Department:
Computer Science
Role:
Creator
More by this author/creator
Institution:
University of Oxford
Division:
MPLS
Department:
Computer Science
Role:
Creator


More from this funder
Funder identifier:
http://dx.doi.org/10.13039/501100000266
Funding agency for:
Turner, H
Georgiev, M
Lovisotto, G
Grant:
EP/P00881X/1
EP/N509711/1
More from this funder
Funder identifier:
http://dx.doi.org/10.13039/501100015278


Publisher:
University of Oxford
Publication date:
2021


Language:
English
Keywords:
Subjects:
Pubs id:
1480632
Local pid:
pubs:1480632
Deposit date:
2021-05-10
ARK identifier:

Terms of use


Views and Downloads






If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP