An Intelligent Music Teaching Quality Evaluation Method Based on Biotechnology-Driven Data and Bio-Inspired Optimization Algorithms

Rong Wang
Zhumadian Preschool Education College, China

Abstract:

As the demand for real-time and accurate evaluation of both instructional processes and student states continues to grow in music education, traditional assessment methods—characterized by strong subjectivity and limited scope—are increasingly inadequate for modern teaching quality management. This study proposes an evaluation framework that integrates biologically driven data acquisition with a bio-inspired optimization strategy. The system collects multidimensional data in music classroom settings, including student behavioral signals, audio characteristics, and physiological engagement indicators, forming a rich, high-dimensional input space. Principal Component Analysis (PCA) is employed to compress the feature space while retaining over 90% of the cumulative variance, thereby effectively reducing redundancy and multicollinearity. For model construction, a bio-inspired swarm intelligence algorithm—the Black-winged Kite Algorithm (BKA), which mimics the hunting behavior of raptors—is used to jointly optimize the input weights and biases of an Extreme Learning Machine (ELM), enhancing its prediction accuracy and generalization capability in nonlinear regression modeling tasks. Empirical results demonstrate that the proposed PCA-BKA-ELM model significantly outperforms conventional methods, including standard ELM, backpropagation (BP) neural networks, support vector regression (SVR), convolutional neural networks (CNN), and bidirectional long short-term memory (BiLSTM) networks in predicting both student performance scores and engagement levels. The model exhibits superior stability, accuracy, and adaptability, offering a lightweight and efficient solution for intelligent, biologically driven evaluation in music teaching scenarios. This approach provides a practical pathway toward real-time, interpretable, and scalable educational assessment systems aligned with the development of smart learning environments.