Behavior-Aware Mobile Sensing for User State Estimation
Project Description
Smartphones play a central role in daily life, yet prolonged and unregulated use can lead to both physical fatigue (e.g., finger or hand strain) and mental fatigue (e.g., reduced attention and alertness). Current digital wellbeing tools, however, primarily rely on simplistic, time-based metrics that do not reflect how the device is used. This project seeks to develop a more intelligent, behavior-driven approach to estimating user fatigue by analyzing interaction patterns directly from smartphone usage.
We will investigate how touch dynamics (e.g., typing speed, tap rhythm, error rates), device motion (e.g., tremor, grip variation), and contextual factors (e.g., app switching frequency, time of day) can serve as implicit indicators of fatigue. Leveraging built-in smartphone sensors—and optionally lightweight external devices such as smartwatches—we will collect user interaction data alongside self-reported fatigue levels. This data will be used to train machine learning models capable of passively inferring fatigue states. The long-term objective is to support user wellbeing by identifying potential overuse or strain based on behavioral quality, contributing to next-generation digital wellbeing systems.
We will investigate how touch dynamics (e.g., typing speed, tap rhythm, error rates), device motion (e.g., tremor, grip variation), and contextual factors (e.g., app switching frequency, time of day) can serve as implicit indicators of fatigue. Leveraging built-in smartphone sensors—and optionally lightweight external devices such as smartwatches—we will collect user interaction data alongside self-reported fatigue levels. This data will be used to train machine learning models capable of passively inferring fatigue states. The long-term objective is to support user wellbeing by identifying potential overuse or strain based on behavioral quality, contributing to next-generation digital wellbeing systems.
Supervisor
SHAO, Qijia
Quota
2
Course type
UROP1100
UROP2100
UROP3100
UROP4100
Applicant's Roles
The applicant will be actively involved in the full lifecycle of the project. Primary responsibilities include:
Assisting in the design and implementation of a smartphone-based data collection tool (e.g., Android/iOS app).
Participating in data collection and management, including pilot testing with users and ensuring data quality and privacy.
Extracting features from behavioral and sensor data (e.g., typing speed, tap frequency, accelerometer patterns).
Supporting the development and evaluation of machine learning models for fatigue or user state estimation.
Contributing to the design and prototyping of fatigue-aware user interface feedback.
Potential paper writing and submission.
Assisting in the design and implementation of a smartphone-based data collection tool (e.g., Android/iOS app).
Participating in data collection and management, including pilot testing with users and ensuring data quality and privacy.
Extracting features from behavioral and sensor data (e.g., typing speed, tap frequency, accelerometer patterns).
Supporting the development and evaluation of machine learning models for fatigue or user state estimation.
Contributing to the design and prototyping of fatigue-aware user interface feedback.
Potential paper writing and submission.
Applicant's Learning Objectives
By participating in this project, the applicant will:
Gain hands-on experience in human-computer interaction (HCI) research and mobile sensing technologies.
Learn to design, implement, and deploy mobile data collection tools for behavioral research.
Develop skills in processing and analyzing sensor data from smartphones.
Apply machine learning techniques to real-world behavioral datasets.
Understand user-centric design principles in the context of digital wellbeing and adaptive interfaces.
Improve communication and collaboration skills through participation in a multidisciplinary research environment.
Gain hands-on experience in human-computer interaction (HCI) research and mobile sensing technologies.
Learn to design, implement, and deploy mobile data collection tools for behavioral research.
Develop skills in processing and analyzing sensor data from smartphones.
Apply machine learning techniques to real-world behavioral datasets.
Understand user-centric design principles in the context of digital wellbeing and adaptive interfaces.
Improve communication and collaboration skills through participation in a multidisciplinary research environment.
Complexity of the project
Moderate