论文标题
音乐创作样式分布的动态群集结构和预测建模
Dynamic cluster structure and predictive modelling of music creation style distributions
论文作者
论文摘要
我们研究音乐创作风格分布的动态，以了解涉及先进智能的文化进化。使用统计建模方法和从日本和美国创建的流行音乐数据集提取的几种音乐统计学（美国），我们探索了集群结构的动态，并构建了一个基于健身的进化模型，以分析和预测音乐创造风格分布的演变。我们发现，集群内动力学，例如集群的收缩和聚类中心的转移，以及由簇相对频率代表的集群间动力学，通常表现出明显的动力模式，这些动力模式在文化和不同的音乐方面都具有。此外，我们发现结合这些动力学模式的进化模型可有效预测未来的创建样式分布，并预测了群集频率和群集差异通常具有可比的贡献。我们的结果突出了群集内动力学在文化进化中的相关性，这些动态在先前的研究中经常被忽略。
We investigate the dynamics of music creation style distributions to understand cultural evolution involving advanced intelligence. Using statistical modelling methods and several musical statistics extracted from datasets of popular music created in Japan and the United States (the US), we explored the dynamics of cluster structures and constructed a fitness-based evolutionary model to analyze and predict the evolution of music creation style distributions. We found that intra-cluster dynamics, such as the contraction of a cluster and the shift of a cluster centre, as well as inter-cluster dynamics represented by clusters' relative frequencies, often exhibit notable dynamical modes that hold across the cultures and different musical aspects. Additionally, we found that the evolutionary model incorporating these dynamical modes is effective for predicting the future creation style distributions, with the predictions of cluster frequencies and cluster variances often having comparable contributions. Our results highlight the relevance of intra-cluster dynamics in cultural evolution, which have often been overlooked in previous studies.
