MALLOW-AWESOME'007

ABSTRACT AND PDF OF ACCEPTED PAPERS AND INVITED TALKS

Munindar Singh, North Carolina State University, USA,
Commitment-Based SOA (Joint work with Amit K. Chopra and Nirmit Desai)

ABSTRACT: The vision of service-oriented computing is centered on business services. By contrast, existing service-oriented architectures are formulated in terms of low-level abstractions that are far removed from business services. This talk describes a new architecture whose components are business services and whose interconnections are modeled in terms of the commitments that support key aspects of service engagements. This talk also shows how this architecture relates to existing SOAs.

BIO: Dr. Munindar P. Singh is a professor in the department of computer science at North Carolina State University. From 1989 through 1995, he was with the Microelectronics and Computer Technology Corporation (MCC). Munindar's research interests include multiagent systems and service-oriented computing, wherein he specifically addresses the challenges of trust, service discovery, and business processes and protocols in large-scale open environments.
Munindar is widely published and cited. Munindar's 1994 book Multiagent Systems, the first book on that subject, was published by Springer-Verlag. He coedited Readings in Agents, which was published by Morgan Kaufmann in 1998. Munindar edited the Practical Handbook of Internet Computing published by Chapman & Hall / CRC Press in October 2004 and coauthored a new text, Service-Oriented Computing: Semantics, Processes, Agents published by Wiley in January 2005.
Munindar cochaired the 2005 edition of AAMAS, the International Joint Conference of Autonomous Agents and MultiAgent Systems. He serves on the founding Board of Directors of the newly synthesized IFAAMAS, the International Foundation of Autonomous Agents and MultiAgent Systems.
Munindar was the editor-in-chief of IEEE Internet Computing from 1999 to 2002. He is a founding member of the editorial boards of the Journal of Autonomous Agents and Multiagent Systems, the Journal of Web Semantics, the International Journal of Agent-Oriented Software Engineering, the Journal of Service-Oriented Computing and Applications, and IEEE Internet Computing.
Munindar's research has been recognized with awards and sponsorship by the National Science Foundation, DARPA, Cisco Systems, Ericsson, IBM, and Intel.
Munindar obtained a B.Tech. in Computer Science and Engineering from the Indian Institute of Technology, Delhi in 1986 and a Ph.D. in Computer Sciences from the University of Texas at Austin in 1993.

HOME PAGE: http://www.csc.ncsu.edu/faculty/mpsingh/
M. Birna van Riemsdijk and Martin Wirsing,
Goal-Oriented and Procedural Service Orchestration - A Formal Comparison

Goals form a declarative description of the desired end result of (part of) an orchestration. A goal-oriented orchestration language is an orchestration language in which these goals are part of the language. The advantage of using goals explicitly in the language is added flexibility in handling failures. In this paper, we investigate how goal-oriented mechanisms for handling failures compare to more standard exception handling mechanisms, by providing a formally defined translation of programs in the goal-oriented orchestration language into programs in the procedural orchestration language, and proving that the procedural orchestration has the same behavior as the goal-oriented orchestration.

Download the PDF of this paper
Douglas M. da Silva and Renata Vieira,
Argonaut: Integrating Jason and Jena for context aware computing based on OWL ontologies (Short paper)

In this paper, we present the integration of the agent-oriented programming framework Jason and the semantic web framework Jena to support ontology-based context aware computing. These technologies together allow for the development of context aware multi-agent systems base on ontologies that describe context.

Download the PDF of this paper
Federico Chesani, Paola Mello, Marco Montali, and Sergio Storari,
Agent Societies and Service Choreographies: a Declarative Approach to Specification and Verification

The need for specifying choreographies when developing service oriented systems recently arose as an important issue. Although declarativeness has been identified as a key feature, several proposed approaches model choreographies by focusing on procedural aspects, e.g. by specifying control and message flows of the interacting services. A similar issue has been addressed in Multi-Agent Systems (MAS), where declarative approaches based on social semantics have been used to capture the nature of agents interaction without over-constraining their behavior. In this paper we show how DecSerFlow can be mapped to SCIFF in an automatic and complete way. DecSerFlow is a graphical language capable to model in an intuitive and declarative fashion service flows, whereas SCIFF is a framework based on abductive logic programming originally developed for dealing with social interactions in MAS. By means of a running example, we show how the conjunct use of both approaches could be fruitfully exploited to declaratively specify and verify service choreographies.

Download the PDF of this paper
Holger Endert, Tobias Küster, Benjamin Hirsch, and Sahin Albayrak,
Mapping BPMN to Agents: An Analysis

In industry the development of software applications is usually a complex and demanding task, and the design and the technical realisation is often spread among different roles, which leads to a time consuming and error-prone exchange of knowledge. In order to ensure the correct translation from business idea to implementation it is crucial to allow for the correct and complete exchange of information between these roles.
In this paper, we describe an automated mapping from business process diagrams to agent concepts that simplify the transfer of knowledge between the roles involved in the software development process. Our approach benefits from building upon an intuitive visual specification language on the one hand, and from using a powerful and flexible execution platform on the other.

Download the PDF of this paper
Guido Boella, Valerio Genovese, Roberto Grenna, and Leon van der Torre,
Roles in Coordination and in Agent Deliberation: A Merger of Concepts

In this paper we generalize and merge two models of roles used in multiagent systems which address complementary aspects: enacting roles and communication among roles in an organization or institution. We do this by proposing a metamodel of roles and specializing the metamodel to fit two existing models. We show how the two approaches can be integrated since they deal with complementary aspects: [1] focuses on roles as a way to specify interactions among agents, and, thus, it emphasizes the public character of roles. [2] focuses instead on how roles are played, and thus it emphasizes the private aspects of roles: how the beliefs and goals of the roles become the beliefs and goals of the agents. The former approach focuses on the dynamics of roles in function of the communication process. The latter approach focuses on the internal dynamics of the agents when they start playing a role or shift the role they are currently playing.

Download the PDF of this paper
Julian Padget, University of Bath, UK,
Regulation Frameworks, Semantics and Service Oriented Architectures

ABSTRACT: The concept of the "institution" as used in the social sciences, management and economics captures the principle of right (and wrong) action and how an observable action in the real world counts as an institutional action, bringing about a change of institutional state. Complementary to these "rules of engagement" is the identification of the right actors - the searcher's problem - and how to describe an actor's attributes effectively - the publisher's problem.
As software development apparently moves towards increasingly open architectures of loosely-coupled software components, we believe institutional models are of increasing relevance as a means to categorize formally the correct and incorrect behaviour of (collections of) software components, that necessarily must operate within multiple regulatory frameworks. Likewise, formal descriptions of institutions and of software components, and the means to reason about them both, appear to have a critical role in supporting the component selection, composition and enactment.
Our work to date has focussed (i) on agent-based systems, developing a formalization of institutions and the interactions between institutions, and (ii) service discovery for semantic web-services, developing a generic matchmaking and brokerage factory framework. The talk aims to draw these strands together and explore how semantic technologies and institutional frameworks might be applied to the construction of service oriented architectures.

HOME PAGE: http://www.cs.bath.ac.uk/~jap/
Simon Wells and Chris Reed,
MAgtALO: Using Agents, Arguments, and the Web to Explore Complex Debates

This paper introduces the MAgtALO system, a prototype environment for online debate that aims to provide a mechanism for supporting naturalistic dialogue. MAgtALO demonstrates how dialogue protocols can be harnessed to achieve two objectives: first, to support flexible intuitive interaction with data in complex, contentious domains in order to facilitate understanding and assimilation; and second, to provide mechanisms for structured knowledge elicitation that allow the resources in those domains to be expanded.

Download the PDF of this paper
Viviana Mascardi, Paolo Rosso, and Valentina Cordì,
Enhancing Communication inside Multi-Agent Systems - An Approach based on Alignment via Upper Ontologies

This paper deals with a theoretical issue related to multi-agent system development and deployment, namely the need of a mechanism for aligning ontologies owned by agents, in order to allow them to communicate in a profitable way. Our approach exploits upper ontologies, i.e., ontologies which describe very general concepts that are the same across all domains, as a “lingua franca” among agents. This approach may overcome some problems that arise in various real scenarios, such as the impossibility for (or the lack of will of) an agent to disclose its own entire ontology to another agent, despite the need to communicate with it. In this paper we propose a comparison of seven existing upper ontologies, and an algorithm for aligning any two (or more) ontologies by exploiting an upper ontology as a bridge.

Download the PDF of this paper
Matteo Baldoni, Cristina Baroglio, Ingo Brunkhorst, Elisa Marengo, and Viviana Patti,
A Service-Oriented Approach for Curriculum Planning and Validation

We present a service-oriented personalization system, set in an educational framework, based on a semantic annotation of courses, given at a knowledge level (what the course teaches, what is requested to know for attending it in a profitable way). The system supports users in building personalized curricula, formalized by means of an action theory. It is also possible to verify the compliance of curricula w.r.t. a model, expressing constraints at a knowledge level. For what concerns the first task, classical planning techniques are adopted, which take into account both the student’s initial knowledge and her learning goal. Instead, curricula validation is done against a model, formalized as a set of temporal constraints. We have developed a prototype of the planning and validation services, by using -as reasoning engines- SWI-Prolog and the SPIN model checker. Such services will be supplied and combined as plug-and-play personalization services in the Personal Reader framework.
Download the PDF of this paper
Giovanni Casella and Vincenzo Deufemia,
Integrating Agents, Ontologies, and Web Services to Build Flexible Sketch-based Applications

We present an approach based on web services, for building open and dynamic agent societies aimed at hand-drawn sketch recognition. The approach exploits ontologies to enable agents to agree on mes- sage semantics and service purposes, standard web services languages to represent agent interaction protocols in a suitable way to be exchanged and handled by agents and web services to expose low-level recognition services. The communication mechanisms that characterize our approach, as well as the modular architecture allow agent societies to self-organize at run time, for gaining the capability of recognizing new domain languages, thus obtaining new exible sketch-based applications.

Download the PDF of this paper
Mauricio Paletta and Pilar Herero,
Extending the FIPA Interoperability to Prevent Cooperative Banking Frauds

Electronic bank transactions are very common today. Services given by an Automatic Teller Machine (ATM), for example, are very popular and widely used by bank clients. Unfortunately, in the same way as the use of these devices is increasing, the proliferation of different frauds to try to violate these systems to steal user's money is also increasing. Sometimes, the modus operandi used by the delinquents depends on different factors, such as the country or the city where fraud is committed or, as in the case of ATMs, the model or location of these devices. Since the detection of these modus operandi is not easy and they could be different from a bank institution to another, having both an environment capable of following up the swindler agents learning processes and a way to prevent the cooperation between these agents to share the learned knowledge, would be very useful to discover different modus operandi before crimes are committed. In this paper, a framework designed to follow up the swindlers' agents learning process and to share the knowledge between the agents is presented. This framework is based on the FIPA (Foundation for Intelligent Physical Agents) specifications and it emphasizes on the swindler agents learning process to fulfil the human-like agent behaviour and a realistic interaction with the environment.

Download the PDF of this paper
The full proceedings of the workshop are available here (PDF, 4.3 MB)