Business Anomaly Detection Based on QoS Benchmark of Resource-service Chain in Collaborative Task

To detect business anomaly caused by resource services in collaborative task, an approach based on QoS benchmark of resource-service chain is presented. Firstly, taking QoS of resource services as feature, the QoS benchmarks are obtained by clustering the relationships between the features of resource services in resource service chains. Then, thresholds of QoS benchmarks are calculated. A dynamic calibration strategy to the threshold is used to perform online detection of business anomaly in global and local views of a collaborative task. Finally, experiments and case studies show that the proposed approach has a high accuracy rate for business anomaly detection.