Research On Anomaly Detection Based On FCM With Adaptive Artificial Fish Swarm

Anomaly detection has been widely used in many fields such as telecommunications,insurance,bank,disaster weather forecasting,and medical fields.It is particularly active especially in the fields of computer graphics,network security,anomaly detection research based on data mining,machine learning and other intelligent technology.Machine learning based on anomaly detection methods requires a large number of samples,especially abnormal samples.Then,based on these abnormal samples,training the rule sets through some intelligent learning method or model,and finally using the trained rule set for abnormal analysis and detection is a hotspot in this research field.The anomaly detection method based on cluster analysis is a typical unsupervised learning technology,which could directly train data on unidentified data sets and establish an anomaly detection model.However,fuzzy C-means(FCM)has its shortcomings,and its application in anomaly detection is not satisfactoryBased on the research background above,this paper presents a network anomaly detection research based on FCM to improve the detection rate of anomaly detection system and reduce the false positive rate of anomaly detection system.FCM is sensitive to the initial value selection and is easy to fall into the local maximum.The detection effect of the anomaly detection algorithm based on it is also not ideal.Therefore,by introducing an artificial fish swarm algorithm with strong global search ability and adding adaptive mechanisms to adaptively adjust the range of visual values,the local and global optimization ability of artificial fish swarm algorithms can be improved,and the number of algorithm iterations can be reduced.Then apply it to the FCM.The FCM clustering analysis is carried out by using the optimal solution obtained by the adaptive artificial f ish swarm algorithm to solve the problems of FCM above.Finally,an anomaly detection algorithm based on adaptive artificial fish swarm FCM is designed to make full use of the advantages of adaptive artificial fish swarm to improve the detection performance of anomaly detection algorithm.Experiments show that the algorithm shows a good level of detection performance on the basis of improving the efficiency of data detection.It provides an effective solution to solve the problems related to the detection rate and false alarm rate in the anomaly detection model.The FCM algorithm reduces the dependence on the initial value by adapting the global and local search capabilities of the artificial fish swarm.The algorithm achieves the purpose of improving the real-time performance of the system by reducing the amount of calculation of the system during the detection process.It is significant for the practical application of anomaly detection.