Research On Operation Situation Early Warning Based On Network Business Fuzzy Correlation Analysis

As the society and technology develops by leaps and bounds,the Internet has become a neural hub that we rely heavily on.The scale of the network and business has also expanded,and the network has become more diverse and complex,making traditional network management work face more difficult challenges.How to ensure the normal,efficient and stable operation of the network and scientifically and accurately predict the trend of network status changes is an urgent problem to be solved in today's network management.Therefore,technologies related to network operation situation have emerged at the historic moment.The network operation situation refers to the current state and changing trend of the entire network composed of factors such as the operating conditions of various network equipment and business systems,network behavior,and user behavior.This thesis focuses on network business,combines big data and fuzzy theory,and conducts research around network operation situation prediction and early warning.This thesis focuses on how to design an intuitive and effective early-warning framework,how to construct comprehensive and comprehensive measurement indicators,and how to predict the network change trend scientifically and accurately.The main contents are as follows:(1)An early warning framework for network operation situation is designed.Based on a hierarchical management structure,this paper proposes a business-based early warning framework that distinguishes the role of each level from the bottom up,making network management more logical and hierarchical.Among them,the business layer is combined with the network fault,and a method that affects the business set by fault solution is proposed,so that the network layer and the business layer are perfectly connected.(2)A comprehensive network operation situation measurement index is constructed.In view of the fact that the damage to some network equipment in the current network does not cause the abnormal operation of network services,this paper focuses on the characteristics of network services,and proposes a network operation situation measurement index two-tuple <actual loss of business,potential business risk>,so that The index consideration is more comprehensive and the meaning is more visual.(3)Improved fuzzy time series prediction model.In order to solve the problem of unsatisfactory and poor interpretability of the fuzzy interval in the prediction model,this paper proposes an improved method for the initial clustering center of the traditional fuzzy C-means clustering algorithm that is sensitive and sets the optimal number of clusters in advance.The class effect is more accurate and the adaptability is improved.Aiming at the problem of ignoring the small membership when extracting fuzzy information in the prediction model,this paper proposes a prediction method based on the second-order Markov model,which can retain all the information of the membership,and can also extract the feature information in the sample sequence to improve prediction accuracy.