Related Experiment Video
Updated: Nov 5, 2025
08:15
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
Published on: March 28, 2025
894
Implicit detection of user handedness in touchscreen devices through interaction analysis.
Carla Fernández1, Martin Gonzalez-Rodriguez1, Daniel Fernandez-Lanvin1
1Department of Computer Science, University of Oviedo, Oviedo, Asturias, Spain.
Peerj. Computer Science
|May 14, 2021
Summary
This study introduces a machine learning approach to automatically detect the hand used to operate a smartphone. This method enhances e-commerce usability by adapting interfaces without draining battery life or requiring device-specific sensors.
Area of Science:
- Human-Computer Interaction
- Machine Learning
- Mobile Computing
Background:
- Mobile devices are increasingly used for web surfing and e-commerce.
- Larger smartphone screens make single-handed operation challenging.
- E-commerce applications need to adapt to user handedness for better interaction.
Purpose of the Study:
- To develop an automatic operating hand detection system for smartphones.
- To improve e-commerce interface adaptability for user handedness.
- To avoid reliance on mobile sensors that impact battery life or calibration.
Main Methods:
- A supervised machine learning classifier was developed.
- Features were extracted from touch traces, including scrolls and button clicks.
- A dataset of 174 users was utilized for training and evaluation.
Main Results:
- The classifier accurately labels the operating hand as left or right.
- The approach demonstrated improved user categorization accuracy.
- The method avoids the use of battery-intensive or calibration-sensitive mobile sensors.
Conclusions:
- Automatic hand detection is feasible using touch trace data.
- This non-platform-specific method enhances mobile e-commerce usability.
- The approach offers a battery-efficient and broadly applicable solution for handedness detection.
