Apparatus and method of semantic service correlation system

• US 9,552,389 B2
• Filed: 07/26/2013
• Issued: 01/24/2017
• Est. Priority Date: 07/09/2008
• Status: Active Grant

• Current Assignee: Maplebear Inc.
• Original Assignee: International Business Machines Corporation
• Inventors: Chieu, Trieu C., Lei, Hui, Xie, Guo Tong, Zeng, Liangzhao
• Primary Examiner(s): HICKS, MICHAEL J
• Application Number:
  US13/951,537
• Publication Number:
  US 20130332448A1
• Time in Patent Office: 1,278 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2453 Query optimisation
• CPC Class Codes: G06F 16/60 of audio data
• CPC Class Codes: G06F 9/5038 considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration scheduling strategies G06F9/4881 and subgroups
• CPC Class Codes: H04L 65/1016 IP multimedia subsystem [IMS]
• CPC Class Codes: H04L 67/51 Discovery or management thereof, e.g. service location protocol [SLP] or web services

Abstract

A correlation-based service mediator consists of Semantic Query Generator, Service Query Manager, Service Invocation Flow Generator, Service Invocation Manager, Expression Interpreter and Service Invocation Result Cache. The Semantic Query Generator converts service request to Semantic Query based on input/output parameters in the service request. The Service Query Manager interfaces with an Ontology Engine, passing query results to Service Invocation Flow Generator, which creates service invocation flow definition based on passed query results. The Service Invocation Manager executes flow definitions. It also manages a service invocation result cache, so that invocation results can be saved and reused for later service request. When the execution results are not available in cache, the Service Invocation Manager invokes service according to flow definition. When attribute dependent functions are invoked Expression Interpreter compute the execution results. The service mediator possesses two major functionalities: service matching and service invocation.

Claims

1. A computer implemented method of service correlation, comprising the steps, performed by a computer in a semantic service mediation system, of:

receiving a service request that provides one or more input parameters and indicates expected output parameters, wherein the one or more input parameters and the expected output parameters include objects according to an object-oriented ontology, wherein an object is defined as an instance of a class;

searching a service repository of available services to identify a candidate service interface set as being a key-based attribute-dependence correlation service interface set that satisfies the service request received in the receiving step, wherein the candidate service interface set is a plurality of service interfaces corresponding to some of the available services in the service repository, wherein the searching step is performed based only on input and output parameters of service interfaces without describing pre- and post-conditions of individual services, and wherein the searching step includes, for the candidate service interface set,

querying whether for all input parameters for the candidate service interface set, each input parameter is either included in the one or more input parameters provided by the service request or else is provided as an output of a service interface in the candidate service interface set,

querying whether all of the output parameters expected by the service request are provided by the candidate service interface set, including

querying whether one or more output parameters expected by the service request are not provided by any individual service interface of the candidate service interface set, and

querying whether outputs of the plurality of service interfaces contain a complete dependence set for each of the one or more expected output parameters that are not provided by any individual service interface, wherein the complete dependence set is a collection of depended classes used together to arithmetically calculate a value for an output parameter,

querying whether the plurality of service interfaces forming the candidate service interface set each require an input parameter that is an object of a key class, including

selecting, from classes corresponding to the one or more input parameters of the service request, one class for designation as the key class, and

for each service interface in the candidate service interface set, comparing each input parameter of each service interface for a match to the designated key class at least until a match is found or all input parameters of each service interface have been compared,

querying whether each of the plurality of service interfaces forming the candidate service interface set contributes a unique output parameter either for the expected output parameters expected by the service request or for the complete dependence set, and

identifying the candidate service interface set as being the key-based attribute dependence correlation service interface set if and only if all of the querying steps return true,

wherein a service interface indicates one or more input parameters required by a corresponding service and indicates one or more output parameters provided by the corresponding service, and

wherein each service interface of the plurality of service interfaces is alone insufficient to satisfy the service request received in the receiving step;

generating a microflow from the plurality of service interfaces forming the key-based attribute-dependence correlation service interface set; and

executing the microflow generated in the generating step to invoke services from the available services in the service repository to satisfy the service request.

2. The method of claim 1, further comprising steps of translating the service request to a semantic query, and using semantic information in service descriptions of the available services in the service repository and in the semantic query, performing one-to-multiple service matchings.

3. The method of claim 2, wherein the searching step further includes executing the semantic query to locate candidate service providers for the service request.

4. The method of claim 1, wherein the searching step further comprises identifying one or more further candidate service interface sets as correlation interface sets based on one or more of parameter-based correlation and attribute-dependence based correlation;

wherein

service interfaces in parameter-based correlation interface sets are correlatable for any common input parameters, and

service interfaces in attribute-dependence correlation interface sets are correlatable for any common input parameters.

5. The method of claim 4, wherein the searching step, using the parameter-based correlation, further comprises the step of locating candidate service interface sets that satisfy the following criteria:

inputs for all service interfaces in a given candidate service interface set are provided by the one or more input parameters from the service request;

the service interfaces in the given candidate service interface set provide all the output parameters expected by the service request;

any outputs of the service interfaces in the given candidate service interface set are correlatable; and

any individual service interface in a the given candidate service interface set contributes a unique output expected by the service request.

6. The method of claim 4, wherein the searching step, using the attribute-dependence correlation, further comprises the step of locating candidate service interface sets that satisfy the following criteria:

inputs for all service interfaces in a given candidate service interface set are provided by the one or more input parameters from the service request;

for any of the output parameters expected by the service request, either a service interface in the given candidate service interface set provides a semantic compatible output, or outputs of service interfaces in the given candidate service interface set contain a complete dependence set for the output required by the service request;

any outputs in the service interfaces of the given candidate service interface set are correlatable; and

any individual service interface in the given candidate service interface set contributes a unique output expected by the service request or the complete dependence set.

7. The method of claim 1, further comprising a step of supporting one-to-one service matchings.

8. The method of claim 1, wherein the searching step includes identifying additional candidate service interface sets that satisfy the service request, wherein the generating step includes generating multiple microflows for the additional candidate service interface sets, followed by selecting a best quality microflow for satisfying the service request using a quality-driven approach.

9. The method of claim 1, comprising a further step of searching the service repository of available services to identify a second candidate service interface set as being a parameter-based correlation service interface set that satisfies the service request received in the receiving step, wherein the second candidate service interface set is a plurality of service interfaces corresponding to some of the available services in the service repository, and wherein the further searching step includes, for the second candidate service interface set,

i. querying whether for all input parameters for the second candidate service interface set, each input parameter is either included in the one or more input parameters provided by the service request or else is provided as an output of a service interface in the second candidate service interface set,

ii. querying whether all of the output parameters expected by the service request are provided by the second candidate service interface set,

iii. querying whether any outputs in the services interfaces of the second candidate service interface set are correlatable,

iv. querying whether each of the plurality of service interfaces forming the second candidate service interface set either

contributes at least one unique output parameter for the expected output parameters expected by the service request, or

appears in interface sequences in query iii. if some outputs are correlated by interface sequences, and

identifying the second candidate service interface set as being the parameter-based correlation service interface set if and only if all of the four querying steps i. through iv. return true.