<![CDATA[This study aims to identify students' learning patterns on sequences and series using a deep learning approach based on the Long Short-Term Memory (LSTM) algorithm, and to analyze the influence of learning patterns and learning motivation on material mastery. The study uses a quantitative approach with an explanatory design, where data analysis is carried out through LSTM modeling for learning pattern classification and Partial Least Squares-Structural Equation Modeling (PLS-SEM) to test the structural relationships between variables. The results show that LSTM based learning patterns have a positive and significant effect on material mastery (β = 0.294; t = 3.160; p = 0.002), learning motivation has a stronger influence on material mastery (β = 0.672; t = 7.038; p < 0.001), and LSTM-based learning patterns have a very significant effect on learning motivation (β = 0.803; t = 24.065; p < 0.001) which indicates the presence of partial mediation. In conclusion, the integration of deep learning-based learning pattern analysis and motivational factors can strongly explain variations in mastery of sequence and series material in vocational school students, with high model predictive ability (R² = 0.856; Q²predict = 0.693).]]>
Copyrights © 2026
