Complex event processing system and deploying method thereof

Abstract

The invention discloses a complex event processing system based on machine learning and a deploying method of the complex event processing system. The complex event processing system comprises an event detection algorithm managing module, a system feature extracting module, a deployment scheme managing module, a system deploying module, a user feedback acquiring module and an environment feature extracting module. The deploying method comprises the following steps: at first, the event detection algorithm managing module and the system feature extracting module are informed of business rules to be customized and system response requirement according to the business rules customized by a user; and then the system deploying module takes the charge of deploying the scheme selected by the deployment managing module, the user feedback acquiring module distributes user identification for captured system response, the environment feature extracting module acquires the features of the current environment, and the deployment scheme managing module optimizes the current deployment scheme. According to the complex event processing system based on machine learning and the deploying method of the complex event processing system, the deploying method can be optimized according to the deployment environment, and the complex event processing system which meets the demands of the deployment environment can be deployed.

Description

technical field [0001] The invention belongs to the technical field of composite event processing (CEP), and in particular relates to a composite event processing system based on machine learning and capable of extracting potential features of an environment and a deployment method thereof. Background technique [0002] With the continuous expansion of application scale, more and more raw data are generated in fields such as security review, online transactions, intrusion detection, and financial services, which puts forward more requirements for information services such as anomaly detection and real-time monitoring. Usually, an event-driven application architecture is adopted, and an event processing system is deployed between raw data and upper-layer applications (MES, ERP, etc.). The event processing system first simply processes the collected data and converts it into basic events. Then, it extracts corresponding compound events from a large number of basic events by us...

AI Technical Summary

Problems solved by technology

The event model, temporal model, event detection language / rules, and composite event response methods that make up the composite event detection algorithm have not been able to compare the performance independently of the special application environment, and there is a lack of analysis of the characteristics and performance of each component of the composite event processing system. awareness of relationship

Images