Automatic Web Services Composition Using Combining HTN and CSP

Semantic Web services and their composition portend a future where the Web can behave more intelligently for solving general real-life problems. To solve such real-life problems requires a set of appropriate services to be composed via planning, scheduled, and then executed: those are logical services composition, physical composition, and execution. Web service composition is the most difficult aspect and is our focus. In this paper, we propose a framework combining logical composition (using HTN) and physical composition (using CSP) for automatic services composition. The framework covers the entire procedures to deal with a user's request, domain analysis of the request, task flow decision and CSP creation by the planner, and solving the CSP by a distributed CSP solver for intelligent Web service composition. Implementation of the framework and an evaluation of it in terms of various problem solving types are then explained. We also discuss the necessity and synergy effect of the combined HTN and CSP for a framework that automates Web service composition and execution.

[1]  Miguel A. Salido,et al.  Scheduling in a Planning Environment , 2000, PuK.

[2]  Edward P. K. Tsang,et al.  Foundations of constraint satisfaction , 1993, Computation in cognitive science.

[3]  James A. Hendler,et al.  HTN planning for Web Service composition using SHOP2 , 2004, J. Web Semant..

[4]  Roman Barták,et al.  Encoding HTN Planning as a Dynamic CSP , 2005, CP.

[5]  Toru Ishida,et al.  A Constraint-Based Approach to Horizontal Web Service Composition , 2006, International Semantic Web Conference.

[6]  Markus P. J. Fromherz,et al.  Constraint-based scheduling , 2001, Proceedings of the 2001 American Control Conference. (Cat. No.01CH37148).

[7]  James A. Hendler,et al.  The Semantic Web" in Scientific American , 2001 .

[8]  Roman Barták.  Integrating Planning into Production Scheduling : A Formal View , 2004 .

[9]  Makoto Yokoo,et al.  Algorithms for Distributed Constraint Satisfaction: A Review , 2000, Autonomous Agents and Multi-Agent Systems.

[10]  Roman Barták.  On the Boundary of Planning and Scheduling: A Study , 1999 .

[11]  Fahiem Bacchus,et al.  Generalizing GraphPlan by Formulating Planning as a CSP , 2003, IJCAI.

[12]  Koustuv Dasgupta,et al.  A service creation environment based on end to end composition of Web services , 2005, WWW '05.

[13]  Henry A. Kautz,et al.  Constraints and AI planning , 2005, IEEE Intelligent Systems.

[14]  Athman Bouguettaya,et al.  A multilevel composability model for semantic Web services , 2005, IEEE Transactions on Knowledge and Data Engineering.

[15]  Peter Norvig,et al.  Artificial Intelligence: A Modern Approach , 1995 .

[16]  James A. Hendler,et al.  Information Gathering During Planning for Web Service Composition , 2004, SEMWEB.

[17]  Dana S. Nau,et al.  SHOP2: An HTN Planning System , 2003, J. Artif. Intell. Res..