Audio classification based on genetic-fuzzy-rough nearest neighbor clustering algorithm

作者：Yu Xiaoqing^*; Shi Jianhua; Liu Huanhuan; Wan Wanggen

来源：Journal of Computational Information Systems, 2013, 9(23): 9397-9404.

DOI：10.12733/jcis7667

• Audio classification
• Classification algorithm
• Cluster centers
• Multi-classifier
• Nearest neighbor-clustering algorithm
• Nearest neighbors
• Rapid expansion
• Research focus

With the rapid expansion of the audio resources, how to achieve a multi-classifier of high performance for audio resources has become one of the research focuses. But for audio data, not only the output of existing class may be fuzzy, and the data itself, there may be roughness. So, in this paper, we propose a new classification algorithm based on fuzzy-rough nearest neighbor theory and genetic algorithm. In which, genetic algorithm is adopted to generate representative cluster center points to classify the test data, with the help of fuzzy-rough nearest neighbor theory. The new algorithm is applied to audio classification, the results show that it is effective and has good performance, which can achieve the accuracy about 90 percent.

• 出版日期2013
• 单位上海大学