Machine Learning-Based Risk Classification of Cybersecurity Behavior Among Smartphone Users

Author(s)	Marc Jordan Ceballos Saladaga
Country	Philippines
Abstract	Smartphone users face a growing landscape of cybersecurity threats due to behavioral weaknesses such as poor password management, use of public Wi-Fi, and unvalidated application downloads. This work proposes a machine learning framework, with Gradient Boosting at its core, to classify users into Low, Moderate, and High-risk classes based on survey-derived behavioral features. We gathered and preprocessed a dataset of 2,751 respondents and trained and compared three models: Gradient Boosting, Random Forest, and Logistic Regression. Gradient Boosting achieved a balanced performance with an accuracy of 0.74 and outperformed baseline models, yielding the highest recall for high-risk detection. To increase transparency, Explainable AI tools such as SHAP and LIME were integrated, showing key behavioral factors that drive risk classification. Our framework has shown how integrating behavioral analysis with interpretable machine learning enables the delivery of customized cybersecurity insights for smartphone users, thereby enhancing digital safety awareness.
Keywords	Cybersecurity, Smartphone Users, Machine Learning, Gradient Boosting, Risk Classification, SHAP, LIME, Behavioral Analysis, Explainable AI, Awareness Modeling
Field	Computer > Network / Security
Published In	Volume 17, Issue 1, January-March 2026
Published On	2026-03-22

About IJSAT Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us	Message on WhatsApp	+91-9687-182-185	editor@ijsat.org

International Journal on Science and Technology