A Multi-Dimensional College Student Behavioral Dataset for Educational Data Mining and Risk Assessment

This dataset was developed to support research in educational data mining, learning analytics, and artificial intelligence applications in higher education. It contains behavioral and academic information collected from college students to better understand how learning habits, classroom participation, and online engagement influence academic performance and student risk levels.

The dataset includes 1,439 student records and captures multiple dimensions of student activity, including demographic characteristics, academic achievement, attendance behavior, Learning Management System (LMS) usage, assignment completion patterns, forum interaction, and video-learning engagement. These attributes were carefully organized to represent realistic student learning behaviors commonly observed in modern higher education environments.

The primary objective of this dataset is to help researchers, educators, and data scientists explore patterns related to student success, engagement, and academic risk. In particular, the dataset can be used for:

student performance prediction,
academic risk identification,
early intervention systems,
behavioral pattern analysis,
explainable artificial intelligence (XAI),
and educational recommendation systems.

A target variable, risk_level, is included to classify students according to their academic risk status, making the dataset suitable for supervised machine learning and predictive analytics tasks.

The dataset combines both academic indicators (such as GPA, course grades, and attendance) and digital learning behaviors (such as LMS logins, session duration, assignment submission rate, and video completion rate). This combination provides a more holistic representation of student engagement in blended and online learning environments.

Because of its structured format and balanced educational features, the dataset can serve as a valuable benchmark resource for researchers working in:

Educational Data Mining (EDM),
Learning Analytics (LA),
Artificial Intelligence in Education (AIED),
student retention studies,
and smart campus research.

The dataset is provided in CSV format and can be directly used with statistical software, machine learning frameworks, and data visualization tools for further analysis and model development.