Based On Data Mining Of Online Banking Customers Perceive Monitoring And Early Warning System Design

In the rapid development of Internet banking today, in order to guarantee the high availability of online banking system key business, and Internet Banking for complex business system for effective monitoring and risk prevention, to better enhance the customer experience, monitoring and warning system plays a vital role in the online banking system maintenance and Optimization in architecture design. Don't allow no monitoring on-line system, monitoring and early warning system by monitoring data acquisition and real-time processing, triggering the alarm event, showing the performance index, the judgment of online banking system is an important basis for the safe operation.This paper through the research of existing application system deployment architecture, application protocol operation mechanism and principle of TCP/IP protocol, based on the specific goal analysis of each module model, determine the overall design scheme of the monitoring and warning system.This paper presents a machine learning based dynamic baseline method, the standard procedure in accordance with the data mining method, through the acquisition of online banking customers perception of the original data, choose reasonable data characteristics and data preprocessing, mainly adopts the reasonable methods to fill or delete the missing data. In addition, according to different data attributes, the method of data standardization is adopted to deal with the difference of dimension. Through the construction of principal component analysis, data mining model dimension reduction algorithm of least squares support vector linear regression support vector machine and nonlinear regression model, realize the dynamic baseline worth prediction, by gradually adding white noise in different degree, in order to determine the warning threshold, finally tested the dynamic baseline model presented in the paper, verified its feasibility and practical.The system as a bank application performance monitoring subsystem has been running on the line, and running well, played the system's monitoring and early warning function.