Modeling Students Digital Behavior for Analyzing Learning Interest Using a Machine Learning Approach

Abstract

The rapid adoption of smartphones among university students has significantly influenced
patterns of digital behavior, including activities that may be associated with learning
engagement and academic interest. Excessive or inappropriate smartphone use can potentially
interfere with students’ learning activities, while productive digital use may support access to
educational resources and communication. This study aims to model students’ digital behavior
and analyze its relationship with learning interest using a machine learning approach. The
study utilizes data collected from university students, including patterns of smartphone usage,
duration of use, frequency of access, types of applications used, and indicators of learning
interest. The data preprocessing stage includes data cleaning, transformation, normalization,
and feature selection to obtain relevant attributes for modeling. Machine learning techniques
are then applied to identify patterns and relationships between students’ digital behavior and
their learning interest. The developed model is evaluated using appropriate performance
metrics to assess its ability to identify learning-interest patterns based on smartphone usage
behavior. The analysis is expected to provide insights into how different patterns of digital
behavior are associated with students’ learning interest and to identify behavioral
characteristics that may contribute to differences in learning engagement. The findings can
support educational institutions and lecturers in understanding students’ digital habits and
developing more appropriate strategies for promoting productive technology use in academic
environments. Furthermore, the proposed approach demonstrates the potential of machine
learning as a data-driven method for analyzing student digital behavior and supporting
educational decision-making. This study contributes to the development of data-driven
educational research by integrating smartphone usage behavior with learning-interest analysis
through machine learning.
..