Abstract: Abnormal network data are affected by the fuzzy weighted disturbance of the clustering center,which leads to poor clustering of data mining.This paper proposed a data mining algorithm of anomaly network based on fuzzy neural network.Similarity analysis is performed according to the mixed classification attributes of abnormal network data,the numerical and classification attributes of abnormal network data are extracted,and the fuzzy fusion processing of the abnormal network data is carried out by using the joint association rule analysis method.The classification fuzzy set of abnormal network data is constructed based on fuzzy centroid heterogeneity measurement method.In the fuzzy data set,abnormal network data are weighted by mixing and adaptive block matching,and the weak correlation characteristic quantity of abnormal network data is extracted.The extracted features are input into the fuzzy neural network classifier for data classification and recognition,and the optimized mining of abnormal network data is completed.The simulation results show that the proposed method has good data clustering ability for anomaly network data mining.The mining process has strong convergence and anti-interference.

Key words: Abnormal network data, Feature extraction, Fuzzy neural network, Mining