摘要
针对数字音频信号分类问题提出了基于二型模糊集合理论的C均值聚类算法,并在此基础上应用跳跃基因遗传算法对聚类得到的初始模糊模型进行优化,最后采用向量相似性测度准则对优化后的模糊规则集合进行简化,得到最终的模糊分类器模型。与传统的一型模糊集合相比,二型模糊集合可以掌控更多的不确定性信息。基于二型模糊集合理论的C均值聚类算法对样本分布不均匀、结构不规则的样本集的聚类效果更精确。实例仿真结果对比显示,应用二型模糊C均值聚类算法的音频信号分类器比应用一型模糊C均值聚类算法的分类器得到的分类结果更准确。
Type-2 fuzzy C-means clustering algorithm is applied to solve the digital audio signal classification problem, and jumping genes genetic algorithm is used to optimized the initial fuzzy model which is obtained by the clustering algorithm. At last, the optimized fuzzy rule base is simplified by the vector similarity measure, and the fimal fuzzy classifier model is obtained. Compared with conventional type-1 fuzzy sets, type-2 fuzzy sets can handle more uncertain information. Type-2 fuzzy C-means clustering algorithm has more accurate results for sample sets whose samples distribute uneven and structures are irregular. The experiment results illustrate that the audio classifier which is based on the type-2 fuzzy C-means clustering algorithm has more precise results than the classifiers which are based on the type-1 fuzzy C-means clustering algorithm.
出处
《电子科技大学学报》
EI
CAS
CSCD
北大核心
2013年第3期436-441,共6页
Journal of University of Electronic Science and Technology of China
关键词
聚类算法
数字音频信号
模糊C均值
遗传算法
二型模糊集合
clustering algorithm
digital audio signals
fuzzy c-means
genetic algorithm
type-2 fuzzy sets
