Dynamic service quality selection
Olga Georgieva · 2016
This paper presents a new approach for software service selection. It applies clustering analysis on quality metrics data that were accumulated during the service monitoring. The assessment is accomplished according to the comparison of properties of the densest clusters found in each data space. For this purpose specific cluster metrics are accounted for. The approach was implemented in a workable procedure that was experimentally proved by real data of web services' monitoring.