Mining art education learning behavior patterns based on TCN

With the deep integration of artificial intelligence and educational technology, educational data mining has gradually become an important means of improving teaching quality and personalized learning, and the mining of learning behavior patterns in art education has also gained much attention. This paper constructs an art education learning behavior pattern mining model based on TCN, aiming to accurately capture the behavioral sequence characteristics of students during the art learning process to enhance the intelligence level of teaching process analysis. This study builds a multi-dimensional behavioral dataset encompassing painting process data, learning activity logs, and classroom interaction records, and performs data preprocessing and behavioral label annotation. Using the TCN model to model students' learning behavior sequences allows for the identification of different types of learning behavior patterns. The results indicate that TCN demonstrates good accuracy and stability in processing the temporal characteristics of art learning behavior, outperforming traditional recurrent neural network models in both training efficiency and behavior recognition accuracy.