Symmetry in computer-aided music composition system with social network analysis and artificial neural network methods

Abstract

To reduce the cost of music creation, shorten the creation time, help composers create music more easily, and make the style of music works more easily accepted by more people, based on the principle of symmetry, network analysis and ANN (artificial neural network) algorithm are applied to design children's music creation system based on GA (genetic algorithm) and ANN. The system includes human–computer interaction technology, learning mechanism, and distributed architecture. To accurately describe music information and extract different types of music features, the Back-Propagation method is used to establish the music classification model. The system helps musicians to create music in real time, overcomes the contradiction between people's evaluation and music generation speed, and retains the advantages of traditional GA. The results show that, compared with other algorithms, the average accuracy of the computer-aided composition technology based on GA and ANN is 97.5%, the best performance in recall rate and synthesis is 94.35% and 93.134%, respectively, and the algorithm model is effective. Moreover, the simulation results of this algorithm show that it is helpful to improve the composition technology and help composers more easily create music, making the style of music works more acceptable to people. This study can provide theoretical basis and practical significance for the research of computer-aided music creation.