Towards an Automatic Non-Deterministic Web Service Composition - - PowerPoint PPT Presentation

▶
towards an automatic non deterministic web service
SMART_READER_LITE
LIVE PREVIEW

Towards an Automatic Non-Deterministic Web Service Composition - - PowerPoint PPT Presentation

Jan 23, 2023 7 likes •691 views

Towards an Automatic Non-Deterministic Web Service Composition Platform Markou, G. & Refanidis, I. Dept. of Applied Informatics, University of Macedonia, Greece NWeSP 2012 - Poster Session Related Work Literature for Web Service

slide-1
SLIDE 1

Towards an Automatic Non-Deterministic Web Service Composition Platform

Markou, G. & Refanidis, I.

  • Dept. of Applied Informatics, University of Macedonia, Greece

NWeSP 2012 - Poster Session

slide-2
SLIDE 2

Related Work

Literature for Web Service Composition (WSC) suggests that:

  • Most widespread method of choice: AI planning

2/12 NWeSP 2012 - Poster Session

slide-3
SLIDE 3

Related Work

Literature for Web Service Composition (WSC) suggests that:

  • Most widespread method of choice: AI planning
  • Use of standards is diverse:
  • OWL-S / PDDL widely used
  • Other standards are also common(e.g., WSMO)
  • Proprietary formats

2/12 NWeSP 2012 - Poster Session

slide-4
SLIDE 4

Related Work

Literature for Web Service Composition (WSC) suggests that:

  • Most widespread method of choice: AI planning
  • Use of standards is diverse:
  • OWL-S / PDDL widely used
  • Other standards are also common(e.g., WSMO)
  • Proprietary formats
  • Very few cases of online WSC systems
  • Especially, currently active

2/12 NWeSP 2012 - Poster Session

slide-5
SLIDE 5

Related Work

Literature for Web Service Composition (WSC) suggests that:

  • Most widespread method of choice: AI planning
  • Use of standards is diverse:
  • OWL-S / PDDL widely used
  • Other standards are also common(e.g., WSMO)
  • Proprietary formats
  • Very few cases of online WSC systems
  • Especially, currently active
  • Gap in the evaluation process of the current WSC systems
  • Many simply evaluate their methodology on a single case study, without

referring to quantitative criteria

  • There are a few notable exceptions, especially in the past few years

2/12 NWeSP 2012 - Poster Session

slide-6
SLIDE 6

System Goals – Functionalities (1/2)

Our work:

  • Makes use of AI planning as our method of choice

3/12 NWeSP 2012 - Poster Session

slide-7
SLIDE 7

System Goals – Functionalities (1/2)

Our work:

  • Makes use of AI planning as our method of choice
  • Uses OWL-S and PDDL*, the “de facto” WS /Planning standards

3/12

*Specifically, NuPDDL

NWeSP 2012 - Poster Session

slide-8
SLIDE 8

System Goals – Functionalities (1/2)

Our work:

  • Makes use of AI planning as our method of choice
  • Uses OWL-S and PDDL*, the “de facto” WS /Planning standards
  • Able to cope with the domain’s non-determinism

3/12

*Specifically, NuPDDL

NWeSP 2012 - Poster Session

slide-9
SLIDE 9

System Goals – Functionalities (1/2)

Our work:

  • Makes use of AI planning as our method of choice
  • Uses OWL-S and PDDL*, the “de facto” WS /Planning standards
  • Able to cope with the domain’s non-determinism
  • Planner-independent

3/12

*Specifically, NuPDDL

NWeSP 2012 - Poster Session

slide-10
SLIDE 10

System Goals – Functionalities (1/2)

Our work:

  • Makes use of AI planning as our method of choice
  • Uses OWL-S and PDDL*, the “de facto” WS /Planning standards
  • Able to cope with the domain’s non-determinism
  • Planner-independent
  • Rigorous evaluation process based on

3/12

*Specifically, NuPDDL

NWeSP 2012 - Poster Session

slide-11
SLIDE 11

System Goals – Functionalities (1/2)

Our work:

  • Makes use of AI planning as our method of choice
  • Uses OWL-S and PDDL*, the “de facto” WS /Planning standards
  • Able to cope with the domain’s non-determinism
  • Planner-independent
  • Rigorous evaluation process based on
  • Quantitative criteria

3/12

*Specifically, NuPDDL

NWeSP 2012 - Poster Session

slide-12
SLIDE 12

System Goals – Functionalities (1/2)

Our work:

  • Makes use of AI planning as our method of choice
  • Uses OWL-S and PDDL*, the “de facto” WS /Planning standards
  • Able to cope with the domain’s non-determinism
  • Planner-independent
  • Rigorous evaluation process based on
  • Quantitative criteria
  • Use case scenarios

3/12

*Specifically, NuPDDL

NWeSP 2012 - Poster Session

slide-13
SLIDE 13

System Goals – Functionalities (1/2)

Our work:

  • Makes use of AI planning as our method of choice
  • Uses OWL-S and PDDL*, the “de facto” WS /Planning standards
  • Able to cope with the domain’s non-determinism
  • Planner-independent
  • Rigorous evaluation process based on
  • Quantitative criteria
  • Use case scenarios
  • Additional manual WSC module as a comparison

3/12

*Specifically, NuPDDL

NWeSP 2012 - Poster Session

slide-14
SLIDE 14

System Goals – Functionalities (1/2)

Our work:

  • Makes use of AI planning as our method of choice
  • Uses OWL-S and PDDL*, the “de facto” WS /Planning standards
  • Able to cope with the domain’s non-determinism
  • Planner-independent
  • Rigorous evaluation process based on
  • Quantitative criteria
  • Use case scenarios
  • Additional manual WSC module as a comparison
  • Will be the 1st online system supporting various stages of WSC

3/12

*Specifically, NuPDDL

NWeSP 2012 - Poster Session

slide-15
SLIDE 15

System Goals – Functionalities (2/2)

Online (Sub)Modules:

4/12 NWeSP 2012 - Poster Session

slide-16
SLIDE 16

System Goals – Functionalities (2/2)

Online (Sub)Modules:

4/12

Registry / Advertisement / Semantic search of Web Services

NWeSP 2012 - Poster Session

slide-17
SLIDE 17

System Goals – Functionalities (2/2)

Online (Sub)Modules:

4/12

Translator between AI and Web Service languages Registry / Advertisement / Semantic search of Web Services

NWeSP 2012 - Poster Session

slide-18
SLIDE 18

System Goals – Functionalities (2/2)

Online (Sub)Modules:

4/12

Translator between AI and Web Service languages Registry / Advertisement / Semantic search of Web Services Automatic WSC module

NWeSP 2012 - Poster Session

slide-19
SLIDE 19

System Goals – Functionalities (2/2)

Online (Sub)Modules:

4/12

Translator between AI and Web Service languages Registry / Advertisement / Semantic search of Web Services Automatic WSC module Also:

  • Manual WSC

module

  • WS descriptions’

editor

NWeSP 2012 - Poster Session

slide-20
SLIDE 20

Proposed Approach - Application

5/12

Initial screen Based on Google Web Toolkit (GWT):

NWeSP 2012 - Poster Session

slide-21
SLIDE 21

Proposed Approach - Application

5/12

User email registration

NWeSP 2012 - Poster Session

slide-22
SLIDE 22

Proposed Approach - Application

5/12

Web Service details

NWeSP 2012 - Poster Session

slide-23
SLIDE 23

Proposed Approach - Application

5/12

Inline

  • perations

NWeSP 2012 - Poster Session

slide-24
SLIDE 24

Proposed Approach - Application

5/12

Semantic search alternatives

NWeSP 2012 - Poster Session

slide-25
SLIDE 25

Proposed Approach - Application

5/12

XML Editor with syntax highlighting Pre-defined OWL-S templates

NWeSP 2012 - Poster Session

slide-26
SLIDE 26

Registry

  • Core of the application is based on iServe
  • Open platform for publishing and discovering services
  • WSs are semantically described in OWL-S, not Minimal Service Model

6/12 NWeSP 2012 - Poster Session

slide-27
SLIDE 27

Registry

  • Core of the application is based on iServe
  • Open platform for publishing and discovering services
  • WSs are semantically described in OWL-S, not Minimal Service Model
  • Used its web-based application for
  • Browsing
  • Querying
  • Uploading services

6/12 NWeSP 2012 - Poster Session

slide-28
SLIDE 28

Registry

  • Core of the application is based on iServe
  • Open platform for publishing and discovering services
  • WSs are semantically described in OWL-S, not Minimal Service Model
  • Used its web-based application for
  • Browsing
  • Querying
  • Uploading services
  • Populated with version 4.0 of OWL-S Test Collection (OWL-S TC)

6/12 NWeSP 2012 - Poster Session

slide-29
SLIDE 29

Registry

  • Core of the application is based on iServe
  • Open platform for publishing and discovering services
  • WSs are semantically described in OWL-S, not Minimal Service Model
  • Used its web-based application for
  • Browsing
  • Querying
  • Uploading services
  • Populated with version 4.0 of OWL-S Test Collection (OWL-S TC)
  • Made several improvements to its interface and functionality

6/12 NWeSP 2012 - Poster Session

slide-30
SLIDE 30

Translation between AI & WS Languages

  • AI planning language: PDDL, the de facto planning language
  • Specifically, NuPDDL
  • Compatible with PDDL2.1, retaining most of it
  • Capable of modeling non-deterministic action effects (one of, unknown)

7/12 NWeSP 2012 - Poster Session

slide-31
SLIDE 31

Translation between AI & WS Languages

  • AI planning language: PDDL, the de facto planning language
  • Specifically, NuPDDL
  • Compatible with PDDL2.1, retaining most of it
  • Capable of modeling non-deterministic action effects (one of, unknown)
  • WS semantic description language: OWL-S, most widely used
  • Heavily influenced from planning languages, such as PDDL

7/12 NWeSP 2012 - Poster Session

slide-32
SLIDE 32

Translation between AI & WS Languages

  • AI planning language: PDDL, the de facto planning language
  • Specifically, NuPDDL
  • Compatible with PDDL2.1, retaining most of it
  • Capable of modeling non-deterministic action effects (one of, unknown)
  • WS semantic description language: OWL-S, most widely used
  • Heavily influenced from planning languages, such as PDDL
  • Partial mapping from OWL-S to PDDL exists
  • Part of it straightforward
  • Conversion of non-physical knowledge from OWL-S to PDDL problematic

 Introduction of new predicate in PDDL domain?

7/12 NWeSP 2012 - Poster Session

slide-33
SLIDE 33

Automatic WSC Module

  • Future work – Basic directions:
  • WSs may fail to execute correctly, e.g.,
  • Unavailability of an atomic WS involved in the plan
  • Output of its successful execution is not the expected one

8/12 NWeSP 2012 - Poster Session

slide-34
SLIDE 34

Automatic WSC Module

  • Future work – Basic directions:
  • WSs may fail to execute correctly, e.g.,
  • Unavailability of an atomic WS involved in the plan
  • Output of its successful execution is not the expected one
  • Generate plans tackling the most influential and likely contingencies

8/12 NWeSP 2012 - Poster Session

slide-35
SLIDE 35

Automatic WSC Module

  • Future work – Basic directions:
  • WSs may fail to execute correctly, e.g.,
  • Unavailability of an atomic WS involved in the plan
  • Output of its successful execution is not the expected one
  • Generate plans tackling the most influential and likely contingencies
  • Proposed solution based on contingent planning

8/12 NWeSP 2012 - Poster Session

slide-36
SLIDE 36

Automatic WSC Module

  • Future work – Basic directions:
  • WSs may fail to execute correctly, e.g.,
  • Unavailability of an atomic WS involved in the plan
  • Output of its successful execution is not the expected one
  • Generate plans tackling the most influential and likely contingencies
  • Proposed solution based on contingent planning
  • Produce seed plan

8/12 NWeSP 2012 - Poster Session

slide-37
SLIDE 37

Automatic WSC Module

  • Future work – Basic directions:
  • WSs may fail to execute correctly, e.g.,
  • Unavailability of an atomic WS involved in the plan
  • Output of its successful execution is not the expected one
  • Generate plans tackling the most influential and likely contingencies
  • Proposed solution based on contingent planning
  • Produce seed plan
  • Examine for significant/ likely points of failure

8/12 NWeSP 2012 - Poster Session

slide-38
SLIDE 38

Automatic WSC Module

  • Future work – Basic directions:
  • WSs may fail to execute correctly, e.g.,
  • Unavailability of an atomic WS involved in the plan
  • Output of its successful execution is not the expected one
  • Generate plans tackling the most influential and likely contingencies
  • Proposed solution based on contingent planning
  • Produce seed plan
  • Examine for significant/ likely points of failure
  • Add conditional branches

8/12 NWeSP 2012 - Poster Session

slide-39
SLIDE 39

Automatic WSC Module

  • Future work – Basic directions:
  • WSs may fail to execute correctly, e.g.,
  • Unavailability of an atomic WS involved in the plan
  • Output of its successful execution is not the expected one
  • Generate plans tackling the most influential and likely contingencies
  • Proposed solution based on contingent planning
  • Produce seed plan
  • Examine for significant/ likely points of failure
  • Add conditional branches
  • Repeat until we reach a plateau or run out of time

8/12 NWeSP 2012 - Poster Session

slide-40
SLIDE 40

Automatic WSC Module

  • Future work – Basic directions:
  • WSs may fail to execute correctly, e.g.,
  • Unavailability of an atomic WS involved in the plan
  • Output of its successful execution is not the expected one
  • Generate plans tackling the most influential and likely contingencies
  • Proposed solution based on contingent planning
  • Produce seed plan
  • Examine for significant/ likely points of failure
  • Add conditional branches
  • Repeat until we reach a plateau or run out of time

Re-planning module will be integrated

8/12 NWeSP 2012 - Poster Session

slide-41
SLIDE 41

Evaluation

  • Currently based on the manual WSC module
  • Modification of an open source BPMN 2.0 editor (PetalsBPM)

9/12 NWeSP 2012 - Poster Session

slide-42
SLIDE 42

Evaluation

  • Currently based on the manual WSC module
  • Modification of an open source BPMN 2.0 editor (PetalsBPM)
  • Making use of OWL-S TC Semantic Web Services v. 4.0
  • Each scenario uses a subset of descriptions of an OWL-S TC domain
  • Several semantic WS descriptions added/ modified for testing purposes

9/12 NWeSP 2012 - Poster Session

slide-43
SLIDE 43

Evaluation

  • Currently based on the manual WSC module
  • Modification of an open source BPMN 2.0 editor (PetalsBPM)
  • Making use of OWL-S TC Semantic Web Services v. 4.0
  • Each scenario uses a subset of descriptions of an OWL-S TC domain
  • Several semantic WS descriptions added/ modified for testing purposes
  • Each scenario with increasing non-determinism & complexity

9/12 NWeSP 2012 - Poster Session

slide-44
SLIDE 44

Evaluation

  • Currently based on the manual WSC module
  • Modification of an open source BPMN 2.0 editor (PetalsBPM)
  • Making use of OWL-S TC Semantic Web Services v. 4.0
  • Each scenario uses a subset of descriptions of an OWL-S TC domain
  • Several semantic WS descriptions added/ modified for testing purposes
  • Each scenario with increasing non-determinism & complexity
  • Quantitative criteria to be considered:

9/12 NWeSP 2012 - Poster Session

slide-45
SLIDE 45

Evaluation

  • Currently based on the manual WSC module
  • Modification of an open source BPMN 2.0 editor (PetalsBPM)
  • Making use of OWL-S TC Semantic Web Services v. 4.0
  • Each scenario uses a subset of descriptions of an OWL-S TC domain
  • Several semantic WS descriptions added/ modified for testing purposes
  • Each scenario with increasing non-determinism & complexity
  • Quantitative criteria to be considered:
  • Number of Web Services considered for WSC

9/12 NWeSP 2012 - Poster Session

slide-46
SLIDE 46

Evaluation

  • Currently based on the manual WSC module
  • Modification of an open source BPMN 2.0 editor (PetalsBPM)
  • Making use of OWL-S TC Semantic Web Services v. 4.0
  • Each scenario uses a subset of descriptions of an OWL-S TC domain
  • Several semantic WS descriptions added/ modified for testing purposes
  • Each scenario with increasing non-determinism & complexity
  • Quantitative criteria to be considered:
  • Number of Web Services considered for WSC
  • Preprocessing time (parsing of ontologies’ concepts, etc)

9/12 NWeSP 2012 - Poster Session

slide-47
SLIDE 47

Evaluation

  • Currently based on the manual WSC module
  • Modification of an open source BPMN 2.0 editor (PetalsBPM)
  • Making use of OWL-S TC Semantic Web Services v. 4.0
  • Each scenario uses a subset of descriptions of an OWL-S TC domain
  • Several semantic WS descriptions added/ modified for testing purposes
  • Each scenario with increasing non-determinism & complexity
  • Quantitative criteria to be considered:
  • Number of Web Services considered for WSC
  • Preprocessing time (parsing of ontologies’ concepts, etc)
  • Transformation time (of WS domain to PDDL one)

9/12 NWeSP 2012 - Poster Session

slide-48
SLIDE 48

Evaluation

  • Currently based on the manual WSC module
  • Modification of an open source BPMN 2.0 editor (PetalsBPM)
  • Making use of OWL-S TC Semantic Web Services v. 4.0
  • Each scenario uses a subset of descriptions of an OWL-S TC domain
  • Several semantic WS descriptions added/ modified for testing purposes
  • Each scenario with increasing non-determinism & complexity
  • Quantitative criteria to be considered:
  • Number of Web Services considered for WSC
  • Preprocessing time (parsing of ontologies’ concepts, etc)
  • Transformation time (of WS domain to PDDL one)
  • Planning time (to output a successful plan)

9/12 NWeSP 2012 - Poster Session

slide-49
SLIDE 49

Evaluation

  • Currently based on the manual WSC module
  • Modification of an open source BPMN 2.0 editor (PetalsBPM)
  • Making use of OWL-S TC Semantic Web Services v. 4.0
  • Each scenario uses a subset of descriptions of an OWL-S TC domain
  • Several semantic WS descriptions added/ modified for testing purposes
  • Each scenario with increasing non-determinism & complexity
  • Quantitative criteria to be considered:
  • Number of Web Services considered for WSC
  • Preprocessing time (parsing of ontologies’ concepts, etc)
  • Transformation time (of WS domain to PDDL one)
  • Planning time (to output a successful plan)
  • Optimality of the outputted plans

9/12 NWeSP 2012 - Poster Session

slide-50
SLIDE 50

Use Case Scenario 1/3: Movie Database

1) Movie database search using a (partial) movie title

10/12 NWeSP 2012 - Poster Session

slide-51
SLIDE 51

Use Case Scenario 1/3: Movie Database

1) Movie database search using a (partial) movie title

  • Based on Communication domain: 58 semantic WSs in total

10/12 NWeSP 2012 - Poster Session

slide-52
SLIDE 52

Use Case Scenario 1/3: Movie Database

1) Movie database search using a (partial) movie title

  • Based on Communication domain: 58 semantic WSs in total
  • Fully deterministic - Serialized composite WS output

10/12 NWeSP 2012 - Poster Session

slide-53
SLIDE 53

Use Case Scenario 1/3: Movie Database

1) Movie database search using a (partial) movie title

  • Based on Communication domain: 58 semantic WSs in total
  • Fully deterministic - Serialized composite WS output
  • Retrieve comedy films with titles similar to user input

10/12 NWeSP 2012 - Poster Session

slide-54
SLIDE 54

Use Case Scenario 1/3: Movie Database

1) Movie database search using a (partial) movie title

  • Based on Communication domain: 58 semantic WSs in total
  • Fully deterministic - Serialized composite WS output
  • Retrieve comedy films with titles similar to user input
  • Output:
  • Movie titles
  • Pricing information (tax-free price, recommended price, etc)

10/12 NWeSP 2012 - Poster Session

slide-55
SLIDE 55

Use Case Scenario 1/3: Movie Database

10/12

Graphical workflow representing the movie database use case scenario

NWeSP 2012 - Poster Session

slide-56
SLIDE 56

Use Case Scenario 2/3: Online Bookstore

2) Online bookstore product purchase

11/12 NWeSP 2012 - Poster Session

slide-57
SLIDE 57

Use Case Scenario 2/3: Online Bookstore

2) Online bookstore product purchase

  • User wants to purchase a book from an specific online bookstore with a

preferred method of payment

11/12 NWeSP 2012 - Poster Session

slide-58
SLIDE 58

Use Case Scenario 2/3: Online Bookstore

2) Online bookstore product purchase

  • User wants to purchase a book from an specific online bookstore with a

preferred method of payment

  • Preferences regarding method of payment

11/12 NWeSP 2012 - Poster Session

slide-59
SLIDE 59

Use Case Scenario 2/3: Online Bookstore

2) Online bookstore product purchase

  • User wants to purchase a book from an specific online bookstore with a

preferred method of payment

  • Preferences regarding method of payment
  • Based on Education domain: 285 semantic WSs in total

11/12 NWeSP 2012 - Poster Session

slide-60
SLIDE 60

Use Case Scenario 2/3: Online Bookstore

2) Online bookstore product purchase

  • User wants to purchase a book from an specific online bookstore with a

preferred method of payment

  • Preferences regarding method of payment
  • Based on Education domain: 285 semantic WSs in total
  • Non-deterministic output
  • Different outcomes:

11/12 NWeSP 2012 - Poster Session

slide-61
SLIDE 61

Use Case Scenario 2/3: Online Bookstore

2) Online bookstore product purchase

  • User wants to purchase a book from an specific online bookstore with a

preferred method of payment

  • Preferences regarding method of payment
  • Based on Education domain: 285 semantic WSs in total
  • Non-deterministic output
  • Different outcomes:
  • Output if book in stock:

 Addition of book in shopping cart  Purchase with the specified method of payment  Output information regarding it, such as its author

11/12 NWeSP 2012 - Poster Session

slide-62
SLIDE 62

Use Case Scenario 2/3: Online Bookstore

2) Online bookstore product purchase

  • User wants to purchase a book from an specific online bookstore with a

preferred method of payment

  • Preferences regarding method of payment
  • Based on Education domain: 285 semantic WSs in total
  • Non-deterministic output
  • Different outcomes:
  • Output if book in stock:

 Addition of book in shopping cart  Purchase with the specified method of payment  Output information regarding it, such as its author

  • No output if book not in stock

11/12 NWeSP 2012 - Poster Session

slide-63
SLIDE 63

Use Case Scenario 2/3: Online Bookstore

11/12

Graphical workflow representing the online bookstore use case scenario

NWeSP 2012 - Poster Session

slide-64
SLIDE 64

Use Case Scenario 3/3: Camera Search

3) Camera search and purchase

12/12 NWeSP 2012 - Poster Session

slide-65
SLIDE 65

Use Case Scenario 3/3: Camera Search

3) Camera search and purchase

  • User want to purchase an analog SLR model; willing to

settle for other ones if it is not in stock

  • User preference towards specific model
  • More than one sellers exist

12/12 NWeSP 2012 - Poster Session

slide-66
SLIDE 66

Use Case Scenario 3/3: Camera Search

3) Camera search and purchase

  • User want to purchase an analog SLR model; willing to

settle for other ones if it is not in stock

  • User preference towards specific model
  • More than one sellers exist
  • Based on Economy domain: 359 semantic WSs in total

12/12 NWeSP 2012 - Poster Session

slide-67
SLIDE 67

Use Case Scenario 3/3: Camera Search

3) Camera search and purchase

  • User want to purchase an analog SLR model; willing to

settle for other ones if it is not in stock

  • User preference towards specific model
  • More than one sellers exist
  • Based on Economy domain: 359 semantic WSs in total
  • Possible iteration in output Web Service
  • Since more than one sellers exist, if product not available at one, check

another

  • If product not found in any of the sellers, try with other versions (digital

SLR, simple compact cameras)

12/12 NWeSP 2012 - Poster Session

slide-68
SLIDE 68

Use Case Scenario 3/3: Camera Search

12/12

Graphical workflow representing the camera search use case scenario

NWeSP 2012 - Poster Session