Research and application on improved fuzzy clustering algorithm

Clustering problem contains pattern recognition, feature extraction and many other mathematical algorithms, which is widely applied in statistics and industrial field. With the increase in data dimension, traditional clustering method cannot meet the actual needs, so the combination of fuzzy mathematics and neural networks is introduced into cluster theory, which can perfectly solve the complex issues like high dimension, large data volume and relative factors' self-coupling. This paper studies BP fuzzy clustering algorithm, classifies phonetic features based on this algorithm and obtains a clustering model with good robustness and high precision. © 2016 American Scientific Publishers All rights reserved.