Service recommendation method based on service combined backbone network weighted K core analysis

Abstract

The invention discloses a service recommendation method based on service combined backbone network weighted K core analysis. The method comprises the following steps: constructing a service subordinate net according to service combined data; sequentially constructing a service combined net and a service combined backbone net, calculating the weighted core values of nodes, carrying out descending sort on the services corresponding to all the nodes in the service combined backbone net; carrying out secondary descending sort on the services corresponding to the sorted nodes which have same weighted core values and same popularity based on the service popularity; selecting any one possible service corresponding to the nodes which have same weighted value and same popularity; based on the sorting result, recommending the services with the weighted core values ranking tops selected from a service set connected with services. According to the method, the problem that few dependence data and difficulty in acquisition of the existing method can be overcome, and the method adaptability can be improved to a certain degree; the service priority is comprehensively evaluated from two aspects of service popularity and composability.

Description

technical field [0001] The invention relates to a service recommendation method, in particular to a service recommendation method based on weighted k-kernel analysis of a service combination backbone network. Background technique [0002] Service-Oriented Computing (Service-Oriented Computing, SOC) is a new type of computing model, which advocates the construction of applications based on services and their combinations, and brings many conveniences to solve the problem of cross-organization agile application integration in distributed and heterogeneous environments , is a topic of much concern in industry and academia. With the popularity of applications composed of services, the types and quantities of services have increased dramatically, and a service-centric Internet is taking shape. Faced with such a large number of service groups, how to find services that meet user needs has become a bottleneck that affects the further development of service computing. [0003] Rec...

AI Technical Summary

Problems solved by technology

[0004] (1) Existing methods basically rely on pre-collected user preference data, service QoS (Quality of Services) information, etc. However, these data are relatively small and difficult to collect, which limits the versatility of the recommendation method;

[0005] (2) Many existing methods rely on the information in the service WSDL (WebServiceDescriptionLanguage) document, but with the diversification of service types, many new services do not have WSDL description documents;

[0006] (3) Lack of service recommendation method for service combination

Images