Navigating Knowledge Dynamics: Algorithmic Music Recombination, Deep Learning, Blockchain, Economic Knowledge, and Copyright Challenges

• Published in: Journal of the knowledge economy Vol. 16; no. 2; pp. 5884 - 5908
• Main Authors: Zhou, Yue, Huang, Fei
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.06.2025; Springer Nature B.V
• Subjects: Artificial intelligence; Artistic expression; Automation; Blockchain; Copyright; Deep learning; Economic Policy; Economics; Economics and Finance; Entrepreneurship; Innovation Management in Asia; Innovations; Knowledge sharing; Musicians & conductors; R & D/Technology Policy; Neural networks in music; Algorithmic music composition; Blockchain for digital collectibles; Integrated AI framework; Economic knowledge sharing; Artistic expression
• ISSN: 1868-7873; 1868-7865; 1868-7873
• DOI: 10.1007/s13132-023-01700-3

In the contemporary era of the knowledge economy, knowledge has assumed a paramount role in production and daily life. Knowledge-sharing technologies rooted in deep learning and blockchain have emerged as prominent research subjects. Within this context, deep learning (DL) is garnering substantial attention, not only for its traditional applications in prediction, classification, and translation but also as a compelling tool for music generation. However, scaling music generation algorithms to create consistently themed and structured artistic works remains a formidable challenge. To address these challenges, this study introduces a novel approach, the Markov Chain Monte Carlo optimized multilayer perceptron algorithm (MCMC-MPA). The primary objective of the MCMC-MPA method is to push the boundaries of conventional art genres by generating visual and auditory artworks. The study involves collecting piano data, which is preprocessed through z-score normalization. Further refinement is achieved using non-fungible tokens (NFTs) to filter unwanted data. Extensive experiments are conducted with real-world datasets to rigorously assess the performance of this innovative hybrid framework. Evaluation criteria, including pitch accuracy, are employed to gauge the effectiveness of the framework. The proposed method exhibits remarkable performance across various metrics, boasting high scores in melody coherence, listening tests, pitch accuracy, relative frequency, and epoch analysis. These results underline the substantial potential of the MCMC-MPA method in the realm of music generation and artistic creation, facilitating the exploration of new frontiers in art and creativity.