This study aims to present an analysis using Spectrogram Analysis, Correlation Analysis, and Multiple Linear Regression Analysis to examine factors affecting the efficiency of teaching management and the relationship between the frequency of attendance and the duration of attendance on academic achievement. The tools employed in this study include event log data analyzed using Process Mining techniques with the Fuzzy Miner algorithm and Spectrogram Analysis through the Matplotlib Library of Python. The sample group consisted of 247 undergraduate students from a private university in Thailand, selected using a purposive sampling method. The results of the Spectrogram Analysis reveal a clear distinction in the continuity of the learning process between groups with academic achievement above 70% and below 70%. The frequency of attendance is positively correlated with the duration of attendance at a statistical significance level of 0.01, and the frequency of attendance (Beta = 0.537) is a significant factor affecting the efficiency of teaching management at a statistical significance level of 0.01. Therefore, it is possible to integrate techniques of Process Mining, Spectrogram Analysis, Correlation Analysis, and Multiple Linear Regression Analysis to discover and confirm methods for developing teaching processes, improving teaching quality, enhancing students' learning experiences, and driving e-learning systems to achieve better academic outcomes, promoting continuous awareness and learning.
Read full abstract