Analysis of Morning Routine Factors Influencing Work Performance Using Data Mining Techniques

Abstract

This research aims to explore morning routine patterns, analyze the relationships between morning behaviors and work productivity, and identify the key factors influencing work productivity. The study applies data mining techniques under the CRISP-DM framework, which include K-Means Clustering for data grouping, the Apriori algorithm for association rule mining, and a Random Forest model combined with Feature Importance and Permutation Importance techniques for variable importance analysis. The data used in this study is the Morning Routine Productivity Dataset. The analysis reveals that emotional state, sleep duration, and physical exercise are significant factors influencing work productivity. These findings can serve as guidelines for improving morning routines—particularly regarding emotional well-being and adequate sleep—to effectively enhance work productivity. Furthermore, they can be applied to develop daily routine recommendation systems in the future.