Erdem Eser Ekinci

Goal-Oriented Agent Testing Revisited

Today multi-agent systems research is ready to be transferred to the industrial applications. But, testing is one of the most critical processes to increase the acceptability of such systems in industrial settings. In this paper, we introduce a goal-oriented testing approach based on test goal concept. This approach alleges that agent goals are smallest testable units in MASs instead of agents unlike other agent testing approaches and tools proposed previously. Moreover, we introduce a testing tool, called as SEAUnit, that provides necessary infrastructure to support proposed approach.

A planner infrastructure for semantic web enabled agents

Web services and agents are two important software development technologies that are affected from the semantic web innovation. Researches for attuning these topics to the semantic web prepare a ground for integration of them. In this paper, a planner infrastructure 1 is introduced that provides the integration of these two topics on the semantic web ground. Our approach is to support the semantic web service usage during agent planning process. Hence, agent can select required external semantic services to satisfy a goal and can interact with the selected services during the plan execution.

An MAS infrastructure for implementing SWSA based semantic services

The Semantic Web Services Initiative Architecture (SWSA) describes the overall process of semantic service execution in three phases: discovery, engagement and enactment. To accomplish the specified requirements of these phases, it defines a conceptual model which is based on semantic service agents that provide and consume semantic web services and includes architectural and protocol abstractions. In this paper, an MAS infrastructure is defined which fulfills fundamental requirements of SWSAs conceptual model including all its sub-processes. Based on this infrastructure, requirements of a planner module is identified and has been implemented. The developed planner has the capability of executing plans consisting of special tasks for semantic service agents in a way that is described in SWSA. These special tasks are predefined to accomplish the requirements of SWSAs sub-processes and they can be reused in real plans of semantic service agents both as is and as specialized according to domain requirements.

Engineering an MAS platform for semantic service integration based on the SWSA

In this paper, a Multi-Agent System (MAS) platform for semantic service integration based on the Semantic Web Services Initiative Architecture (SWSA) is discussed. We define a software architecture in order to provide concrete realization of the SWSA. The architecture fullfills fundamental requirements of the SWSAs sub-processes. Software agents are employed in automatic discovery and execution of the Semantic Web Services within this architecture. We also elaborate implementation of SWSAs sub-processes (service advertisement, discovery, engagement and enactment) taking the main components of the defined architecture and their interactions into consideration. Hence, the developers can easily utilize semantic web service technologies by using this flexible and extensible platform.

Modeling tools for platform specific design of multi-agent systems

In this paper, we introduce platform specific modeling and code generation tools for the model driven development of multi-agent systems (MAS). These tools enable agent developers to model their MASs for the SEAGENT and the JADEX agent platforms based on the semantics and design principles of these platforms. The toolkit also provides automatic code generation for agent developers in order to implement their MASs on the target platforms. Generated codes may vary on type (e.g. Java class files, XML documents or ontologies) according to each platforms requirements.

Developing multi agent systems on semantic web environment using SEAGENT platform

In this paper, we discuss the development of a multi agent system working on the Semantic Web environment by using a new framework called SEAGENT. SEAGENT is a new agent development framework and platform, which includes built-in features for semantic web based multi agent system development. These features provide semantic supports such as a new specific content language for transferring semantic knowledge, specifically designed agents internal architecture to handle semantic knowledge, a new directory facilitator architecture based on semantic service matching engine and ontology management service to provide ontology translations within the platforms ontologies. The implemented case study shows the effectiveness of these features in terms of semantically enriched multi agent system development.

Exception Handling in Goal-Oriented Multi-Agent Systems

Cooperative, autonomous and distributed properties of multi-agent systems deduce the dynamic capabilities of multi-agent system applications. On the other hand, these suitable features increase the error proneness of these applications. In this paper, we propose an exception handling approach to make multi-agent system applications more reliable and robust. And also we classify multi-agent exceptions and have implemented our approach on SEAGENT goal-oriented multi-agent development framework.

Refactoring in Multi Agent System Development

The need for XP-like agile approaches that provide evolutionary development in a flexible way has been widely acknowledged in the AOSE area. Such approaches improve acceptability of agent-technology by industry. Evolutionary development of multi agent systems-MASs can only be applied successfully, if designs of the MASs being developed are improved throughout the development process. In this paper, we introduce a refactoring approach that can be used during evolutionary MAS development. The proposed refactoring approach makes it possible to develop MASs in an evolutionary way by managing the changes between the iterations of the evolutionary development process. Also, a case study that shows application of a refactoring technique during the evolutionary MAS development is introduced in the fifth section.

Satisfying agent goals by executing different task semantics: HTN, OWL-S or plug one yourself

In this paper, we propose a plan execution architecture which supports different task semantics. This way, each goal that is identified during the deliberation cycle of an agent can be satisfied through tasks defined in different semantics. The capability of supporting different task semantics provides two main advantages. The first advantage is the reuse of legacy artifacts within agent plans. The second advantage is simplifying the adaptation of an agent architecture to different standards of a business organization. In order to integrate various task semantics within a plan execution architecture, we have used a smoothly revised version of a previously articulated workflow model into which different task semantics would be reduced before execution. We have integrated hierarchical task network and OWL-S semantics into our plan execution architecture to test the strength of it in terms of support for executing different task semantics in an agent architecture. We think that such a plan execution architecture will contribute to the industrialization of agent architectures.

Description Logic Based BDI Implementation for Goal-Directed Semantic Agents

The Semantic Web, in its visionary architecture, employs intelligent agents fulfilling the user goals on the web content that is declared with Description Logic (DL) based Semantic Web languages. In order to meet this task, two important points must be taken into account in agent frameworks. First, frameworks must support comprehensive goal models that allow to pursue goals rationally. Second, these goal models must be integrated with the Semantic Web languages to enable defining goals depending on the web content. However, to execute and manage such goal models, elements of the agent architecture must be adapted with respect to the Semantic Web languages and the DL components behind these languages. For this purpose, in this paper, we propose a DL based goal model and introduce a Belief-Desire-Intention (BDI) architecture which is built on top of DL components. In this architecture, we focus on how declarative goals are represented and managed.