Software Agents

In this paper we propose software engineering subontology. We called it application-specific ontology, for specific software development. It enables remote team members browsing, searching, sharing, and authoring ontological data under the distributed software engineering projects environment. We transform explicit meaningful human knowledge into application-specific ontology, where knowledge structures and semantics are linked, and we go through a formal hand-shaking agreement establishing process before the semantic contents are updated in ontology repositories. The application-specific ontology is used for communication over project agreement to facilitate better, highly consistent communications and formalized domain knowledge sharing. We assume that object-oriented development is deployed in the distributed projects. The knowledge of object-oriented development formed in the application-specific ontology clarifies the objectoriented development concepts in a machine understandable form. Software agent, for example, can be utilised to extract information.

The concept of provenance is already well understood in the study of fine art where it refers to the trusted, documented history of some work of art. Given that documented history, the object attains an authority that allows scholars to understand and appreciate its importance and context relative to other works of art. This same concept of provenance may also be applied to data and information generated within a computer system; particularly when the information is subject to regulatory control over an extended period of time. Today's distributed architectures (not only Agent technologies, but also Web Services' and GRID architectures) suffer from limitations, such as lack of mechanisms to trace results. Provenance enables users to trace how a particular result has been arrived at by identifying the individual and aggregated services that produced a particular output. In this chapter we present the main results of the EU PROVENANCE project and how these can be valuable in agent-mediated healthcare applications. For the latter we describe the Organ Transplant Management Application (OTMA), one of the demonstrator applications developed.

New deployment scenarios tend to consider the requirement of session continuity for service provisioning, especially multimedia streaming, to limited heterogeneous portable devices roaming among wireless localities. In particular, multimedia streaming should not experience any interruption while clients roam in wired-wireless integrated networks based on the standard besteffort Internet. The paper proposes an application-level middleware approach to proactively overcome Wi-Fi handoff and maintain multimedia session continuity in the wireless Internet by exploiting mobile proxies running on the wired network. Mobile middleware proxies locally support resource-limited clients, avoid packet losses during handoffs, pre-fetch local buffers with multimedia contents before handoff occurrence, and possibly reconfigure/renegotiate ongoing sessions after handoffs. Experimental results show that, notwithstanding the application-level implementation, mobile proxies can avoid streaming discontinuities with good efficiency in wireless-wired integrated networks even if their pro-activity is based on simple and lightweight handoff prediction techniques.

The decision activity in the field of environmental management is highly complex and involves a great number of contradictory interests (socio-economic, ecological, etc.). Environmental problems often result from the distributed and uncoordinated land use management practices of individual decision-makers that, when taken together, cause significant environmental impacts. To develop feasible and politically acceptable solutions to such problems it is often necessary to foster compromise and consensus among

Negotiation-based transactional mechanisms provide flexibility and economic benefits to both sellers and buyers on online trading platforms. Although automated negotiation is a highly desired feature among online platform providers, the complexity and un- certainty of human behavior in human-to-computer (HtC) negotiation make it a problematic solution. This study proposes a design theory for a human-learning guided virtual negotiator (HLG-VN) framework that emulates human learning using multiple machine learning (ML) techniques that collectively mimic four human learning activities: didactic, feedback, observational, and analogical learning. Fol- lowing the design science research methodology, we built an instantiation system for the proposed design theory and empirically tested it using experiments based on HtC negotiations. The empirical results show that our system outperformed the benchmark system in terms of both economic and some key social-psychological outcomes. Furthermore, the experiment results confirm the effectiveness and cor- rectness of the HLG-VN framework. The proposed design theory provides a theoretical base for using ML techniques to build a virtual negotiator agent for an automated negotiation system. Thus, various agents could be designed and developed based on the theory for online trading platforms, thus improving negotiation efficiency and reducing transaction costs.

In the last year or so (and going back many decades) there has been extensive claims by major computational scientists, engineers, and others that AGI (artificial general intelligence) is 5 or 10 years away, but without a scintilla of scientific evidence, for a broad body of these claims: Computers will become conscious, have a "theory of mind," think and reason, will become more intelligent than humans, and so on. But the claims are science fiction, not science. This article reviews evidence for the following three (3) propositions using an extensive body of scientific research and related sources from the cognitive and neurosciences; evolutionary evidence; linguistics; data science; comparative psychology; self-driving cars, and robotics; and the learning sciences. (1) Do computing machines (LLMs) think or reason? (2) Are computing machines (LLMs) sentient or conscious? (3) Do computing machines (LLMs) have a theory of mind?

Web service technology is one of the trend technologies in computer science now, and recently with the rapid growth of web services deployed with the same functionality, many researches proposed selection systems to automatically rank and select the optimum web service between the deployed candidates. However, the role of these systems limited at selecting the optimal web service only, and they did not introduce a direct solution to step over failure situations in real time systems with out changing the client application. In this work, we introduce a new automated web services supervision system based on Multi Agents System platform named JADE to provide two tasks. First, to select the optimal web service with respect to users' preferences based on changeable Quality of Service (QoS) criteria in dynamic environment; Second, to take advantage of agents' technology to manage, monitor the communication between services and clients, and automatically manipulate connection failures.

In Ambient Intelligence (AmI) environments, some services provided by AmI devices are often not visible to users and to other devices. The existing approaches deal with services' composition and discovery as two independent parts. In this paper, we propose an alternative approach based on logical reasoning agent system. This system is supported by a communication protocol where agents discover automatically services provided in their environment and construct dynamically composite services. The service composition is constructed from an exchange of idiomatic expressions among agents and users, while the discovery process takes the form of an information request via the communication protocol. The advantage of this approach is that agents are able to acquire knowledge from each other and when interacting with users. This capability will facilitate the satisfaction of user's requirements in an intelligent way. This study shows that agents are able to satisfy new services previously unknown to the system.

KM initiatives in organisations have often run into difficulties at the implementation level. Research into the causes of these problems has shifted attention to the human dimension of KM. Recently some researchers have focused on personal knowledge management ...

This paper discusses the use of intelligent Internet agents as a way of expanding audit practices in the virtual world. A quality/service framework is presented that suggests ways that accounting firms can evolve in this era of on-line opportunities. The Edgar Agent is presented as an example of an intelligent Internet agent that gathers financial information. This information can be provided as a value-added service by auditors to their clients and used by the accounting firm itself in performing the audit function.

We study a generalized version of the near-duplicate detection problem which concerns whether a document is a subset of another document. In text-based applications, document containment can be observed in exact-duplicates, near-duplicates, or containments, where the first two are special cases of the third. We introduce a novel method, called CoDet, which focuses particularly on this problem, and compare its performance with four well-known near-duplicate detection methods (DSC, full fingerprinting, I-Match, and SimHash) that are adapted to containment detection. Our method is expandable to different domains, and especially suitable for streaming news. Experimental results show that CoDet effectively and efficiently produces remarkable results in detecting containments.

Collaboration among enterprises in a dynamic environment makes the actors to concentrate on their respective core competences and allow provision and sharing of expertise, resources, and skills for taking advantages and better respond to business opportunities. The coming together of such organizations, usually enhanced by computer network, is referred to as virtual enterprise (VE) or task group. This partnership is only possible if the systems in the various associated organizations can process the data in one another. A major challenge in the enterprise collaborative system that has attracted many research efforts in the recent past is semantic interoperability. This collaborative-based market place requires among others common conceptualization and meaningful data exchange. Effective collaboration among VE members is a major key to the accomplishment of this noble objective. For this interoperability to be effective, we propose in this paper an ontology-based middleware framework...

Agentic AI-autonomous, goal-driven AI agents-can understand high-level tasks and take actions without constant human supervision[1]. Recent large language models (LLMs) such as GPT-5 demonstrate emerging agentic capabilities, autonomously generating user interfaces, backend code, and complete applications from single prompts[2][3]. We survey how these advancements could dramatically reduce the complexity of full-stack software development and democratize coding, enabling non-experts to build end-to-end applications through natural language. We further discuss use cases where agentic AI automates development workflows and everyday activities, as well as challenges (e.g., reliability, scalability, safety) and promising future directions (e.g., improved memory and tool integration) in realizing this vision[4][5].

Recibido el 24 de agosto de 2007; aceptado el 3 de marzo de 2008. Se presenta un sistema multi-agente dedicado a tareas de control de procesos. El objetivo es aplicar la tecnología de agentes inteligentes a una estrategia de control, para que el control sea flexible, autocoordinado y autónomo. También se hace una comparación contra una estrategia de control avanzado para mostrar las ventajas de los agentes de software como controladores. Se escogió una columna de destilación binaria como proceso de prueba debido a las severas interacciones, respuesta no lineal, grandes tiempos muertos y gran tamaño. Este proceso es muy común en las industrias de refinación del petróleo y petroquímica. Las capacidades de los agentes pueden explotarse para obtener una estrategia de control capaz de sobreponerse a estas características del proceso. Como resultado se obtiene una estrategia de control capaz de sobreponerse a las interacciones y adaptable a la respuesta no lineal del proceso, que no se logra con el control convencional. Palabras clave: control inteligente, agentes de software, columna de destilación binaria. A Multi-Agent System dedicated to process control tasks is presented. The objective is to apply the concepts of agent technology to a control strategy to make it flexible, self-coordinated and autonomous. Also a comparison against an advanced control system is made in order to show the advantages of software agents as controllers. A binary distillation column was selected as a test process because of its severe interactions, nonlinear response and long dead times. This process is common in oil refining and petrochemical industries. Agent capabilities can be exploited to obtain a control strategy capable to overcome the process features like interactions and nonlinear response.

Recibido el 24 de agosto de 2007; aceptado el 3 de marzo de 2008. Se presenta un sistema multi-agente dedicado a tareas de control de procesos. El objetivo es aplicar la tecnología de agentes inteligentes a una estrategia de control, para que el control sea flexible, autocoordinado y autónomo. También se hace una comparación contra una estrategia de control avanzado para mostrar las ventajas de los agentes de software como controladores. Se escogió una columna de destilación binaria como proceso de prueba debido a las severas interacciones, respuesta no lineal, grandes tiempos muertos y gran tamaño. Este proceso es muy común en las industrias de refinación del petróleo y petroquímica. Las capacidades de los agentes pueden explotarse para obtener una estrategia de control capaz de sobreponerse a estas características del proceso. Como resultado se obtiene una estrategia de control capaz de sobreponerse a las interacciones y adaptable a la respuesta no lineal del proceso, que no se logra con el control convencional. Palabras clave: control inteligente, agentes de software, columna de destilación binaria. A Multi-Agent System dedicated to process control tasks is presented. The objective is to apply the concepts of agent technology to a control strategy to make it flexible, self-coordinated and autonomous. Also a comparison against an advanced control system is made in order to show the advantages of software agents as controllers. A binary distillation column was selected as a test process because of its severe interactions, nonlinear response and long dead times. This process is common in oil refining and petrochemical industries. Agent capabilities can be exploited to obtain a control strategy capable to overcome the process features like interactions and nonlinear response.

Formation flying is an enabling technique for numerous proposed satellite missions. The TeamAgent software system, based on ObjectAgent technology, is designed to enable multiple satellites to cooperate autonomously with particular focus on formation flying. A generalized agent-based software architecture for formation flying has been devised and is being applied to the Air Force Research Laboratory's TechSat 21 technology demonstration mission.

Agent-based modeling and simulation (ABMS) is an approach to modeling systems comprised of individual, autonomous, interacting "agents." There is much interest in many application problem domains in developing agent-based models. Agent-based modeling offers ways to model individual behaviors and how behaviors affect others in ways that have not been available before. Applications range from modeling agent behavior in supply chains and the stock market, to predicting the success of marketing campaigns and the spread of epidemics, to projecting the future needs of the healthcare system. Progress in the area suggests that ABMS promises to have far-reaching effects on the way that businesses use computers to support decision-making and researchers use agent-based models as electronic laboratories to aid in discovery. This brief tutorial introduces agent-based modeling by describing the basic ideas of ABMS, discussing some applications, and addressing methods for developing agent-based models.

The impact of infusing breakthrough autonomy technology into a flight project was a big surprise. Valuable technical and cultural lessons, many of general applicability when introducing system-level autonomy, have been learned by infusing the Remote Agent (RA) into NASA's Deep Space 1 @S 1 ) spacecraft. The RA's architecture embodies system-level autonomy in three major components: Planning and Scheduling, Execution, and Fault Diagnosis and Reconfiguration. Lessons learned include: The architecture was confirmed. Active participation by nonautonomy personnel in the development is essential. Communication of new concepts is essential, difficult, and hampered by differences in terminology. Giving a spacecraft system-level autonomy changes organizational roles in operating the spacecraft after launch, and hence changes roles during development. Software models supporting functions traditionally handled on the ground must be developed early enough to get on-board. Shortfalls in planned features must be technically and developmentally accomodatable, in particular not to threaten the launch schedule. Traditional commanding must be supported. Testing must be emphasized; end-to-end tests counter skepticism. These lessons and others, on incremental system releases and use of autocode generation, are based on 16 months of spiral development from start of project through the project's decision to reduce the role of the RA from full-time control of the spacecraft to a separable experiment. TABLE: OF CoNrmT's 1. INTRODuCTiON 2. T}IE NEW MILLENNIUM PROGRAM 3. DSI AUTONOMY/FLIGEII' SOFTWARE ARCIIITECTURE 4. FLIGIIT SOFIWARE IMPLEMENTATION APPROACH 5. CIIALLENGING INITIAL CONDITIONS 6. ROLE REDUCTION OF TIIE REMOH; AGEN"I' 7. LESSONS LEARNED 8. SUh4MARY

1MADS Group Universidade da Coru˜na (Spain) 2Babel Group Universidad Politécnica de Madrid (Spain) 3LambdaStream Servicios Interactivos SL Ronda de Outeiro 33 Entlo., A Coru˜na, Spain ... D. Castro Using McErlang to Verify an Erlang Process Supervision Component

Supercapacitors-based storage systems are expected to play a key role in microgrids in view of their capability to compensate high-power imbalances. We define an agent for the control of supercapacitor arrays within the context of the novel control framework Commelec, proposed by the Authors as a composable method for real-time control of active distribution networks with explicit power setpoints. An important function of such an agent is to advertise the real-time power capabilities and operational preferences of the supercapacitor array based on local information. Given the small energy capacity of such a device, its internal state can largely vary from one setpoint implementation to the next one. For this reason, the use of an accurate model is crucial in the agent definition. We show that it is possible to infer the real-time power capabilities of the device by using simple measurements on the supercapacitor array suitably coupled with an accurate representation of the cells composing the array. Results show that the agent is able to speak for the resource, thus allowing its use from an external controller.

algorithmic mechanism design, auction and negotiation technology, automated trading, ZIP traders, genetic algorithms, e-marketplaces This paper builds on previous papers describing our ongoing research in automated market-mechanism design: using a genetic algorithm (GA) to find optimal parametersettings for software-agent traders that operate in virtual "e-marketplaces", where the rules of the marketplaces are also under simultaneous control of the GA. The aim is that the GA automatically designs new agent-based e-marketplaces that are more efficient than existing markets designed by (or populated by) humans. Das et al. recently demonstrated that ZIP software-agent traders consistently out-perform human traders in Continuous Double Auction (CDA) marketplaces similar to those used in the international financial markets. Cliff used a GA to explore a continuous space of ZIP-trader auction-market mechanisms, where the space of possible auction-types explored included the CDA and also two purely one-sided mechanisms. Surprisingly, the GA would sometimes settle on novel hybrid auction mechanisms partway between the CDA and a one-sided auction. Such results occurred when the market's supply and demand schedules were unchanging, and also when the schedules undergo a single sudden "shock" change halfway through the evaluation process. These results could prima facie support the hypothesis that hybrid auctions are in general preferable to the well-known CDA mechanism. In this paper we present new results that clarify that hypothesis. In our new experiments, more than one shock-change in supply/demand occurs during the mechanism-evaluation process; under this regimen, CDA auction-mechanisms are identified by the GA as optimal in three of the four experiments reported here, and in the fourth experiment a near-CDA hybrid mechanism was evolved. From these results we conclude that, while evolved hybrid market-mechanisms may be useful in niches where the marketplace's supply and demand dynamics are known a priori to be relatively stable or regular, the CDA remains the mechanism of choice when it is difficult or impossible to make accurate predictions concerning the dynamic stability of the supply and demand schedules.

algorithmic mechanism design, auction and negotiation technology, automated trading, ZIP traders, genetic algorithms, e-marketplaces This paper builds on previous papers describing our ongoing research in automated market-mechanism design: using a genetic algorithm (GA) to find optimal parametersettings for software-agent traders that operate in virtual "e-marketplaces", where the rules of the marketplaces are also under simultaneous control of the GA. The aim is that the GA automatically designs new agent-based e-marketplaces that are more efficient than existing markets designed by (or populated by) humans. Das et al. recently demonstrated that ZIP software-agent traders consistently out-perform human traders in Continuous Double Auction (CDA) marketplaces similar to those used in the international financial markets. Cliff used a GA to explore a continuous space of ZIP-trader auction-market mechanisms, where the space of possible auction-types explored included the CDA and also two purely one-sided mechanisms. Surprisingly, the GA would sometimes settle on novel hybrid auction mechanisms partway between the CDA and a one-sided auction. Such results occurred when the market's supply and demand schedules were unchanging, and also when the schedules undergo a single sudden "shock" change halfway through the evaluation process. These results could prima facie support the hypothesis that hybrid auctions are in general preferable to the well-known CDA mechanism. In this paper we present new results that clarify that hypothesis. In our new experiments, more than one shock-change in supply/demand occurs during the mechanism-evaluation process; under this regimen, CDA auction-mechanisms are identified by the GA as optimal in three of the four experiments reported here, and in the fourth experiment a near-CDA hybrid mechanism was evolved. From these results we conclude that, while evolved hybrid market-mechanisms may be useful in niches where the marketplace's supply and demand dynamics are known a priori to be relatively stable or regular, the CDA remains the mechanism of choice when it is difficult or impossible to make accurate predictions concerning the dynamic stability of the supply and demand schedules.

Gode and Sunder 1993 described continuous double-auction cda markets populated by zerointelligence" zi traders that act randomly. Their results appear to indicate that no intelligence is required to give h uman-like trading performance. Cli and Bruten 1997 demonstrated serious failings of zi traders. Here, `zi-plus' zip traders are introduced: simple stochastic agents that adapt over time using an elementary form of machine learning. It is shown that the performance of zip traders is signi cantly closer to the human data than is that of zi traders. Thus, human-like trading behavior can be achieved with intelligence that is more than zero but much less than human.

The "ZIP" adaptive trading algorithm has been demonstrated to outperform human traders in experimental studies of continuous double auction (CDA) markets. The original ZIP algorithm requires the values of eight control parameters to be set correctly. A new extension of the ZIP algorithm, called ZIP60, requires the values of 60 parameters to be set correctly. ZIP60 is shown here to produce significantly better results than the original ZIP (called "ZIP8" hereafter). A genetic algorithm (GA) is used to search the 60-dimensional ZIP60 parameter space, and it finds parameter vectors that yield ZIP60 traders with mean scores significantly better than those of ZIP8s. This paper shows that this optimizing evolutionary search works best when the GA itself controls the dimensionality of the search-space, so that the search commences in an 8-d space and thereafter the dimensionality of the searchspace is gradually increased by the GA until it is exploring a 60-d space. Furthermore, the results from ZIP60 cast some doubt on prior ZIP8 results concerning the evolution of new 'hybrid' auction mechanisms that appeared to be better than the CDA.

Gode and Sunder's (1993) results from using "zero-intelligence" (zi) traders, that act randomly within a structured market, appear to imply that convergence to the theoretical equilibrium price in continuous double-auction markets is determined more by market structure than by the intelligence of the traders in that market. However, it is demonstrated here that the average transaction prices of zi traders can vary significantly from the theoretical equilibrium value when the market supply and demand are asymmetric, and that the degree of difference from equilibrium is predictable from a priori probabilistic analysis. In this sense, it is shown here that Gode and Sunder's results are artefacts of their experimental regime. Following this, `zero-intelligence-plus' (zip) traders are introduced: like zi traders, these simple agents make stochastic bids. Unlike zi traders, they employ an elementary form of machine learning. Groups of zip traders interacting in exper...

A previous CIFEr paper demonstrated that a genetic algorithm could be used to automatically discover new optimal auction mechanisms for automated electronic marketplaces populated by software-agent traders. Significantly, the new auction mechanisms are often unlike traditional mechanisms designed by humans for human traders; rather, they are peculiar hybrid mixtures of established styles of mechanism. This previous work concentrated on auction marketplaces populated by software agents running the ZIP trader algorithm (recently shown to outperform human traders). In this paper we provide the first demonstration that qualitatively similar results (i.e., non-standard hybrid mechanism designs being optimal) are also given when similar experiments are performed using a different trader algorithm, namely Gode & Sunder's ZI-C traders. Thus, this paper is the first to show that the previous results were not specific to ZIP traders, and hence it offers significant evidence that evolved hybrid auction mechanisms may be found to outperform traditional mechanisms for any style of trader-agent.

In this paper we propose an architecture for the development of Intelligent Virtual Environments for Training, which is based on a collection of cooperative software agents. The first level of the architecture is an extension of the classical Intelligent Tutoring System architecture that adds to the expert, student, tutoring and communication modules a new module which is called World Module. Several software agents compose each module. Moreover, the proposed architecture includes agents able to simulate the behavior of human students and tutors, as well as agents able to plan the procedures to be taught (given an initial state and a desired final state) prior to the tutoring process.

In this paper, we describe EMMA (E-Mail Management Assistant), an e-mail system that addresses the process of e-mail management, from initially sorting messages into virtual folders, to prioritizing, reading, replying (automatically or semi-automatically), archiving and deleting mail items. EMMA attains a high degree of accuracy on email classification by using a rule-based approach known as Ripple Down Rules (RDR) as the basis of rule construction. In contrast to traditional rule-based systems, RDR systems provide extensive help to the user in defining rules and maintaining the consistency of a rule base, making EMMA easy to use. We discuss the results of evaluating the usability of EMMA on a trial of independent users.

Sustainability has progressively become significant to research and practice in educational sector over the year as a result of rapid reduction of natural resources and sudden climatic changes. The fast changing global weather requires Information Technology (IT) practitioners in university institutions to respond to these changes and implement Green strategies and initiatives. However the literature on implementing Green practices by university campuses is still in its infancy, an emphasis on sustainability has been argued to support IT practitioners enhance the use of IT infrastructures, while achieving a sustained university campus. While sustainability has gained increasing attention from researchers and academicians in organizations and industries, sustainability practice implementation in university sector is still not fully successful since IT practitioners currently uses manual based Green assessment methods. IT practitioners in university campuses lack the capability to assess their current Green practices automatically. Therefore there is need for architecture to support IT practitioners in assessing, benchmarking and rating their current Green practice implementation to ascertain if the sustainability goal of economic, people and environment is archived or if they need to improve their Green process implementation. To overcome the aforementioned problems this research proposes an agent based Green assessment architecture to evaluate the current practice implementation in university campuses automatically. The proposed Green assessment system architecture also provides Green process implementation suggestions to IT practitioners.

The International Competition and Exhibition on Computing Innovation (iCE-CInno 2016) is an international research competition organised by Universiti Malaysia Pahang. The competition aims to promote the sprit of inventiveness among Malaysians and ASEAN participants with the theme "Communitising Innovative Computing". This competition is a platform for young inventors to introduce and exhibit their creations that optimistically will drive technologies in future.

Green Information technology system offers opportunity for IT based organizations to act proactively in terms of environmental preservation as well as to mitigate effects of global climate change and other environmental problems. Green Information technology system adoption and implementation is a plausible attempt for IT based organizations to resolve the current environmental issues and also enhance their economic performance and possible social benefits. Currently several models and frameworks have been developed to address sustainability in various research domain. However, there are few models that can support decision making of practitioners in adopting and implementing sustainable practices in IT based organization. Therefore there is need to develop a model to assist practitioners adopt and implement sustainable practices. The proposed model comprises of Green variables, Green process and a hybrid technique of software agents and case bases reasoning. Findings from this pape...

The change in climatic conditions has resulted to the decline of the environment and excess consumption of resources have put more pressure on IT practitioners to Green their business process in collaborative enterprise. Thus there is need to reduce environmental impacts of enterprise development life cycle. IT practitioners however, still face challenges in diffusion Green practices in their enterprise. They often implement new projects with different project requirements and characteristics. Therefore, it would be helpful if IT practitioners could refer to other similar Green cases when planning and implementing current projects. Thus this paper proposes a case based reasoning decision support model to provide case studies of similar Green cases projects and suggestions on how to diffuse sustainable practices in collaborative enterprise. The model can support IT practitioners in making decisions on Green practices that can be diffused in their enterprise process. This paper synthesis and extract secondary data from existing Green techniques and available Green tools. Our findings show that no tools or techniques have been proposed to cater these issues in collaborative enterprise domain. Thus the proposed model is a valuable tool for IT practitioners in collaborative enterprise.

This paper presents a novel agent-based individuals from different communities of interest framework as a decision aid tool for intelligence analysis. represented by agents. This technology extends net-centric information processing and abstraction as well as fusion and multi-source Owing to their flexibility and dynamic nature, intelligent integration strategies. Our information agents traverse and agents are widely used as an interface between users and the mediate disparate ontologies in different formats providing a Internet applications. For example, Bollacker utilized an foundation for semantic interoperability. The presented agent that assists the user in searching for scientific system provides knowledge discovery by accessing a large literatures [3], Ackerman developed I-DIAG agent system number of information sources in a particular domain and to refine group collective discussions to be more condensed organizing them into a network of information agents. Each and strong knowledge [1], and Lieberman employed an agent provides expertise on a specific topic by drawing on agent for helping users browse the Internet by retrieving relevant information from other information agents in relevant documents according to users' browsing activities related knowledge domains. Unique advantages include net- [13]. Other attempts such as an information agent called centric scalability, principled information assurance, as well SurfAgent to automatically recommend relevant documents as ground breaking knowledge discovery in service of to users based on their profiles [18], or an agent-based intelligence analysis. brokering facilitation between users and various information resources utilize agents to assist users in retrieving

Agent-Oriented Software Engineering (AOSE) poses several challenges to the traditional theory and practice of software engineering with the emergence of a variety of theories and tools proposed to deal with this challenge. Agent frameworks with formal foundations have an important role to play in this scenario by allowing analysis based on a mathematical model. It is our contention that the available frameworks and languages do not fulfil all aspects of agent-based system analysis and development. We perform a survey of the available formal theories and tools for Multi-Agent Systems (MAS) and derive a strong motivation for a new formal framework to support AOSE. Several aspects of agent behaviour in this framework have been inspired by the popular Belief-Desire-Intention (BDI) model of cognitive agents. The new framework introduced concisely in the second half tackles the issues raised in the first half, including those regarding mobility, proactive behaviour, organisation and inter-agent communication.

In this paper, we present a negotiation framework in which the negotiation process is performed at two levels. The upper level deals with the formation of high-level goals and objectives for the agent, and the decision about whether or not to negotiate with other agents to achieve particular goals or bring about particular objectives. Negotiation at this upper level determines the rough scope of the commitment. The lower level deals with feasibility and implementation operations; negotiation at this level involves refinement of the rough commitments proposed at the upper level. The experimental work shows this two-level negotiation framework enables the agent to handle complicated negotiation issues and uncertainties in a more efficient way.

dvances in information retrieval (IR) technologies have led to the development of such tools as AltaVista and Google, which give information seekers a starting point for their searches. However, in many cases, manual browsing through even a limited portion of the relevant information is no longer effective. The problem originates not in the IR technologies, but rather in the volumes of information available via the Web and the generality of the term-frequency approach to searches. For any given query, there are often simply too many relevant documents for the human client to process the information efficiently. Over the past five years, we have been developing a Web-based, resource-bounded, information-gathering agent-called BIG, for Bounded Information Gathering-to support a human decision process. Although its techniques are general enough to apply to a wide range of domains, BIG specifically helps clients pick software packages. For example, a client can instruct BIG to recommend a database package for Windows 98 and specify constraints on both the search process and the product characteristics. BIG will then formulate a plan, locate and extract relevant information from both structured and unstructured documents, and return a recommendation to the client, along with supporting data. The BIG system and its performance are well documented. 1,2 We give a brief overview of how it works in the sidebar on p. 51, "BIG Agent Architecture and Information-Gathering Scenario," but our focus here is on a recently developed capability, namely, to schedule information-gathering activity in a way that controls the money spent on acquiring information from sites that charge a fee for access. This capability supports an "information marketplace" on the Web, which we feel will eventually supersede (but obviously not completely replace) the current model of free information access supported by advertising fees.

Trends in distance education show a growing emphasis in collaborative learning, stimulating students to exchange ideas and information. A collaborative environment, however, will demand a higher effort from the teacher, who will have to supervise all the discussions among the learners, so that they do not deviate from the intended topic for the lesson. Moreover, the information proceeding from the interactions among the students will provide to the teacher features that allow a individual evaluation of the students and his course. In this way, this paper describes a first experience using Multi-agent architecture able to monitor the communication tools in a distance learning group. This system analyzes the discussions taking place in these tools (discussion list, chat and newsgroups), showing to the teacher statistical information (percentile of participation and number of changed messages) and identifying possible associations in the interactions, such as, topics and subtopics that interest the students, groups of learners that interact intensively, etc.

The aim of this paper is to combine standard planning and replanning methods into a rigorous unifying framework, extending an existing logic-based approach to resource-based planning. In this Action Resource Framework (ARF), actions and resources are the primitive concepts. Actions consume and produce resources. Plans are structured objects composed of actions and resource schemes and an explicit dependency function specifying their interrelationships. Previous plans can be used both for creating new plans and for modifying plans. Since we are often interested in reusing only a part of these previous plans, we extend the Action Resource Formalism with incomplete plans. To eciently represent such incomplete plans we use the notion of a gap. We maintain a library of incomplete plans (with gaps), and we present operators to insert plan parts from this plan library into the current plan. We prove this set of plan operators to be complete. Generalizing the renement planning approach, we present a template algorithm for both planning and replanning using the ARF and its plan library and plan operators. Finally, we show that existing (re)planning methods and heuristics nicely t into this framework.

Trajectory simulation is a software module that computes the flight path and flight parameters of munitions. It is used throughout the engineering process, including simulations for studying the design trade-offs, to mission simulations for defended area analysis. In this wide application domain, reuse has always been one of the challenges of the trajectory simulation community. We apply an ontology-based simulation development methodology to fulfil the functional requirements of a trajectory simulation while targeting reuse through interoperability and composability. Trajectory Simulation ONTology (TSONT) has been constructed as a simulation conceptual model for trajectory simulations. Based on the knowledge captured in TSONT, a domain-oriented reuse methodology has been leveraged to develop HLA-compliant trajectory simulations. A trajectory simulation federate was developed by conforming to the simulation object model based on TSONT. This paper demonstrates our approach to achieve composable and interoperable simulations over a case study in which a trajectory simulation federate serves in a variety of federations that have been constructed.

SciBox is an end-to-end automated science planning and commanding system. Planning starts with science objectives expressed as algorithms, and ends with commands validated against resource constraints and health and safety rules, ready for uploading to a spacecraft. The process is largely automated, with user capabilities to edit the end sequence if needed. The immediate benefits of an automated system are improved operations efficiency and flexibility and reduction in operations cost through a reduction in manpower effort.

features Antonio Camurri, Paolo Coletta, Barbara Mazzarino, Stefano Piana DELIVERABLE D4.2 Grant Agreement no. 289021 Project acronym ASC-Inclusion Project title Integrated Internet-Based Environment for Social Inclusion of Children with Autism Spectrum Conditions Contractual date of delivery 31 October 2012 Actual date of delivery 31 October 2012 Deliverable number D4.2 Deliverable title Implementation of EyesWeb low-level features Type Report, Public Number of pages 15 WP contributing to the deliverable WP 4 (Nonverbal Expressive Gesture Analysis) Responsible for task Antonio Camurri (UNIGE) antonio.camurri@unige.it

Recent advances in generative AI have increasingly highlighted the transformative potential of large language models (LLMs) within educational contexts. Nevertheless, the token-based generation characteristic of LLMs often results in responses that may lack depth, thereby potentially limiting their effectiveness as scaffolding tools. This paper introduces Alice, a scaffolding agent designed to provide unsolicited hints and adaptive support in computational thinking (CT) education. Alice's effectiveness was primarily evaluated through user feedback scores and benchmarked programming tasks, while further empirical research is underway to explore qualitative evidence of scaffolding effectiveness. Alice was initially optimized for both plugged and unplugged learning scenarios using a structured system prompt informed by a hierarchical framework for AI-mediated CT and the oDSP-HF approach. Subsequent fine-tuning with a LoRA-based method reduced perplexity from 9.5 to 6.6 and improved JavaScript-to-block-based code conversion accuracy from 45.8% to 69.5%. User ratings also increased from 64% to 85%. These findings tentatively indicate that careful system prompt design, combined with targeted fine-tuning, may enhance the adaptive support and learner engagement provided by LLM-based scaffolding agents in CT education.