School of Music, Xi'an University, Xi'an, Shannxi 710065, China.
Comput Intell Neurosci. 2022 May 30;2022:5601689. doi: 10.1155/2022/5601689. eCollection 2022.
Electronic music can help people alleviate the pressure in life and work. It is a way to express people's emotional needs. With the increase of the types and quantity of electronic music, the traditional electronic music classification and emotional analysis cannot meet people's more and more detailed emotional needs. Therefore, this study proposes the emotion analysis of electronic music based on the PSO-BP neural network and data analysis, optimizes the BP neural network through the PSO algorithm, and extracts and analyzes the emotional characteristics of electronic music combined with data analysis. The experimental results show that compared with BP neural network, PSO-BP neural network has a faster convergence speed and better optimal individual fitness value and can provide more stable operating conditions for later training and testing. The electronic music emotion analysis model based on PSO-BP neural network can reduce the error rate of electronic music lyrics text emotion classification and identify and analyze electronic music emotion with high accuracy, which is closer to the actual results and meets the expected requirements.
电子音乐可以帮助人们缓解生活和工作中的压力，是人们表达情感需求的一种方式。随着电子音乐类型和数量的增加，传统的电子音乐分类和情感分析已经不能满足人们越来越详细的情感需求。因此，本研究提出了基于 PSO-BP 神经网络和数据分析的电子音乐情感分析，通过 PSO 算法对 BP 神经网络进行优化，结合数据分析提取和分析电子音乐的情感特征。实验结果表明，与 BP 神经网络相比，PSO-BP 神经网络具有更快的收敛速度和更好的最优个体适应度值，能够为后续的训练和测试提供更稳定的运行条件。基于 PSO-BP 神经网络的电子音乐情感分析模型可以降低电子音乐歌词文本情感分类的错误率，准确识别和分析电子音乐情感，更接近实际结果，满足预期要求。
