Research And Implement Of Financial Network Service Quality Monitoring System Based On Deep Learning

With the rapid development of network technology,the scale of financial services has increased year by year,and the network systems of financial institutions such as banks have become more and more complex.The diversity of services seriously affects the operation of core services in financial networks and reduces the performance of network services.In such a complex financial network where multiple services coexist,different types of services have great requirements on network performance.For example,transaction-type services,as a key business of banks,must be fully guaranteed.In order to provide different quality of service to different services,it is urgent to monitor and analyze the quality of service of financial networks.At present,the network monitoring system basically measures the quality of service from the network layer.Although it meets some limited requirements,it lacks the monitoring of service quality at the service level.The premise of monitoring the quality of service from the service layer is to identify the type of service to which the traffic belongs.However,with the enhancement of network security awareness,financial institutions began to encrypt the traffic of transaction services.Traditional traffic identification methods,such as port matching and deep packet inspection,have failed,while the accuracy of machine learning methods based on stream features cannot be Meet the requirements of quality of service monitoring.In this paper,the financial network is deeply researched,and the financial encryption service traffic classification method based on deep learning is designed,and the financial network service quality monitoring index system is constructed to realize the real-time monitoring of financial network service quality from the network and service layers.The specific research results of this paper include:1.In order to solve the classification problem of cryptographic service traffic in financial network,analyze and compare various encryption traffic classification methods,based on the excellent performance obtained by deep learning technology,use convolutional neural network to construct financial network encrypted traffic classification model.The experimental results show that the model achieves excellent performance on the encrypted traffic dataset in the financial network when the model training is optimal.The classification accuracy,recall rate and F1 value can reach 98%.2.Through in-depth analysis of the monitoring requirements of the service quality of transaction services in the financial network,establish a service quality indicator system for the financial network,and explain the calculation method of the service quality monitoring indicators of the transaction service,so as to achieve the purpose of ensuring the customer's key business experience.Finally,the deep learning-based financial network service quality monitoring system proposed in this paper is deployed in the actual financial network environment,and the service quality monitoring results are presented and analyzed through multiple dimensions to guide network optimization and improve service quality.The test results show that the system can better identify the encrypted traffic and monitor the operation of the financial network in real time and comprehensively.