ABSTRACT In this paper, we propose a context‐aware fragrance recommendation model, which moves the task beyond static similar scent matching towards situational recommendation, where different perfumes are appropriate for different social environments. In our formulation, a perfume collection is modelled as a dynamic olfactory wardrobe, where each perfume is tailored to specific environmental and social demands. We formalize each perfume through a unified representation which combines contextual suitability and affective signals extracted from user reviews, together with olfactory catalogue data. The proposed model integrates deterministic top‐k ranking and stochastic candidate selection to balance relevance and exploration. We assess the performance of the model with complementary evaluation measures covering identity retrieval, ranking quality of success hits, semantic utility and catalogue exploration. The results indicate that while even simple popularity baselines remain highly competitive in raw retrieval precision, our fused context‐affective model provides stronger contextual utility than single‐feature strategies. Furthermore, our stochastic ranking variant substantially improves catalogue coverage with only a limited loss in retrieval quality.

Modern applications increasingly rely on knowledge and collective wisdom. While the internet offers abundant sources, it is untrustworthy, incomplete, and essentially a vast, heterogeneous database lacking current information. With innovation accelerating exponentially, harnessing collective intelligence is essential to drive progress. We present Consensus Without Authority, a general, ledger-backed meta-protocol framework that transforms any group of self-interested actors into a trusted collective-intelligence engine. Each actor i produces a local insight $L{I_i} = {f_1}\left( {\mathcal{Q},{\theta _i}} \right)$, evaluates its peers Ui= f2({LIj}), and enters a two-phase commit-reveal cycle. Harmonizers aggregate votes V = f3({Ui}), synthesize candidate knowledge CKj= f4(V,{LIi}), and a smart contract finalizes each round r by CK(r)= f5({CKj}). We argue that, under majority-honest assumptions and token-weighted incentives, honest play is a Nash equilibrium and the framework converges to a unique fixed point even in Byzantine settings.While first applied to federated learning, the framework is not limited to distributed ML training. It supports socially scoped cognition, where actors contribute local knowledge and reasoning to a shared space, enriching common knowledge iteratively.Two case studies illustrate its range: (i) crowd-sourced RLHF, where dispersed annotators guide policy updates without sharing labels; (ii) federated LoRA tuning of large language models across siloed hospitals, achieving near-centralized accuracy under HIPAA and GDPR.We place the framework within an emerging "AI-native stack": a Byzantine ledger for immutable state, Google’s A2A for agent-to-agent messaging, and Anthropic’s MCP for secure tool access — together enabling a Cognitive Internet where autonomous agents learn, trade, and verify knowledge without centralized trust, creating a scalable, interoperable substrate for next-generation human–AI collaboration.

In this paper, we propose a method for obtaining personalized book recommendations using content-based filtering approach and three sets of book features to define the book metadata, in order to highlight their impact in the recommendation process. The recommender system is experimentally validated using four books datasets of different sizes, collected from Goodreads website - a popular book social network, using our customized web scraper. Lastly, we propose three evaluation metrics: Coverage, Average Recommendations Similarity and Relevance, and discuss our results.

The Internet of Things (IoT) paradigm is evolving and the Next-Generation IoT (NG-IoT) ecosystem will incorporate distributed ledger and blockchain technology, AI-adapted components, and intelligent edge solutions that take advantage of edge computing, Artificial Intelligence (AI), networks, and communications. In addition to the low integration of eXplainable Artificial Intelligence (XAI) in the IoT or NG-IoT contexts, the explainability of these systems is rarely evaluated. Due to these limitations, we thoroughly examined the current state of XAI integration with IoT services. We propose a new conceptual framework called eXING-IoT (eXplainability Integrated in the Next Generation IoT) for better NG-IoT systems' explainability integration and evaluation. This includes a list of qualities that future NG-IoT environments should have, thus paving the way for the advancement of NG-IoT beyond the state of the art.

In this contribution, we present a method for obtaining literature books recommendations using collaborative filtering recommender system technique and emotions extracted from multi-source online reviews.We experimentally validated the proposed system using a book dataset and associated reviews that we collected from Goodreads and Amazon websites using our customized web scrapers.We show the benefits of using multisource reviews by proposing a series of recommender system evaluation measures, which include single-source and multisource recommendations similarity, recommendation algorithm usecases coverage and generated recommendations relevance.

The current surge in AI trends has catalyzed a strong inclination among organizations to transition towards AI-driven solutions. However, a significant challenge arises from the prevalent monolithic nature of existing applications, which often impedes scalability and limits the potential for enhancement through agent-based interventions. This paper aims to investigate strategies for transitioning from monolithic applications to microservices-based architectures and explore the utilization of agents for control within microservices environments. Subsequently, drawing from existing literature and our own insights, we endeavor to formulate a comprehensive strategy for transforming original monolithic applications into intelligently controlled microservices-based systems. We conclude with an IoT use case in order to illustrate the application of this strategy and highlight the advantages that can be achieved.

There is great potential in leveraging Artificial Intelligence (AI) systems to optimize complex infrastructures, automate difficult tasks, or support autonomy and coordination between networked devices. However, advances in state-of-the-art AI often neglect features and/or requirements that businesses care deeply about, namely traceability and explainability. While majority of research concerning Explainable AI remains focused on weight modelling and timid gray-box approaches, the state-of-the-art has not explored much the deployment of semi-physical architectures combining fuzzy rule-based systems with more opaque models to improve explainability. This contribution aims to explore and make the case for a middle ground of mixed AI architectures that combine the performance of black-box AI models with a more explainable overall architecture, enabling operators to use them, while still retaining the core aspects of explainability, when compared to full black-box AI systems. This work contextualizes a potential application of such approach to the problem of Service Level Agreement compliance, in a case of microservice allocation decision over cloud (and cloud-like) infrastructures.

In this contribution, we propose a method for providing User-Based Collaborative Filtering Recommendations using the emotions present in social media reviews. We use a deep learning model which identifies the review dominant emotion from the 6 primary emotions proposed by W.G. Parrott. For experiments, we use a dataset containing book reviews and movie reviews that we collected from Goodreads and IMDB websites using our customized web scrapers. Moreover, we represent the book and associated movie as an unique item in the dataset. We validated our recommender system using a set of system-unseen reviews that simulate a set of users seeking for recommendations. The top k recommendations received for each simulated user are then analyzed by our proposed performance measures.

The measurement of aircraft noise is very important for the industry of air transportation, for residents and for the municipality.In the United States of America, the Federal Aviation Authority (FAA) has a managing tool for aircraft noise map called Aeronautical Environmental Design Tool (AEDT) (Boeker Fleming, 2008); in French Bruitparif is the tool which satisfies the European directive for noise map elaboration strategy. The noise map enables city authorities, residents, and the airport manager to control the noise produced by aeronautical activities in order to master, for instance, the building of infrastructures surrounding the airport zone - a dataset built with a relational data base management system obtained from secondary radar detection. These data, after having gone through a data cleaning process, will be brought and tested with functions such as Naive Bayes, decision tree and random forest. Based on confusion matrix as a metric, we assessed the provided result by the decision tree with metrics as “ 98% for precision, 97% for recall and 97% for f1-score”. This article emphasizes that on the one hand, the data used for machine learning have the same quality in terms of integrity and precision of the radar itself and on the other hand the performance resides on the confusion matrix.

Network engineers are essential to the management of computer systems, as they configure services and devices on the network to guarantee effective data routing and improve computer networks which make easier to perform everyday tasks like sending and receiving emails as well as more sophisticated ones like cloud computing and online gaming. In addition, modern networks have to serve a rising number of IoT devices without compromising performance and handle more sophisticated cybersecurity threats. Network engineers must always be learning and adapting to new protocols and technologies if they are to successfully tackle these difficulties. Moreover, the introduction of Large Language Models (LLMs) that are available as open-source software has revolutionized technical innovation by enabling the automation of network setups and augmenting the capabilities of network administration. These developments represent an important step forward in the field of network engineering, with the goal of maximizing efficiency and guaranteeing strong network security and functioning.

Finding a job these days is challenging because of the size, diversity, and goals of the market in a society impacted by pandemics, economic crises, or military hostilities. Trust is the most crucial factor in the job domain after performance expectations. It is particularly significant for women, less active job seekers, and people who did not experience job recommendations. Since recommender systems (RS) are one of the most frequently encountered human-centered and online applications in our daily lives, it is important to note that sound principles of trusting the environment of Artificial Intelligence (AI) systems are also required to characterize the trustworthiness of recommender systems. Otherwise, inadequate advice, high expectations and bad interpretations could lead to making bad choices or to demotivating job seekers. This paper expands on previous research, highlighting the point of view of trustworthiness in job recommender systems (JRS) and providing an overview of the dimensions of AI trustworthiness for the job domain. The purpose of this study is to investigate how trustworthy and suggestive outputs can improve the communication between a job mediator and a job seeker by enhancing the credibility of the information provided to job applicants and increasing customer satisfaction.

In this paper we propose two book recommendation methods based on emotions extracted from user reviews, using content-based filtering and collaborative filtering.The methods were experimentally evaluated on our own dataset that we collected from Goodreads -a popular website with large database of books and readers reviews.We created an experimental setup where the recommendation algorithms for carrying out the evaluation using two proposed evaluation metrics: coverage and average recommendations similarity.

Current society is facing remarkable changes in last decade directed by several processes: gains in life expectancy leading to ageing population; essential transformation caused by different epidemic situations including the recent COVID-19 pandemic; on-line work and isolation that leads to different levels of mental disorders; stressful working environments and everyday life. All these circumstances generate significant risk factors that drastically influence peoples' bad health status and conditions.On the other hand, in a lot of, even highly developed countries, health and medical support is getting more and more limited and out of normal working conditions due to different reasons.So, a very important question in this highly technologically developed society is if and how information communication technologies, artificial intelligence, and virtual agent technologies can help in changing and improving the current situation.One important scientific and research direction is oriented towards the development of high quality and reliable personalized medical and health services. The development of sophisticated, intelligent virtual agents i.e. a specific kind of medical e-coaching can help people in getting adequate advices and recommendations in order to improve their health conditions and quality of key life indicators.The use of agents in personalized and social medical and health platforms can open new possibilities for producing tailored recommendations during human-intelligent virtual agents' conversation and support. Some opportunities and challenges of human-agent communication are considered in order to increase human living and health conditions. Several paradigmatic cases of intelligent virtual agents are presented and challenges for future development.

Reusing ontologies in practice is still very challenging, especially when multiple ontologies are (jointly) involved. Moreover, despite recent advances, the realization of systematic ontology quality assurance remains a difficult problem. In this work, the quality of thirty biomedical ontologies, and the Computer Science Ontology are investigated, from the perspective of a practical use case. Special scrutiny is given to cross-ontology references, which are vital for combining ontologies. Diverse methods to detect potential issues are proposed, including natural language processing and network analysis. Moreover, several suggestions for improving ontologies and their quality assurance processes are presented. It is argued that while the advancing automatic tools for ontology quality assurance are crucial for ontology improvement, they will not solve the problem entirely. It is ontology reuse that is the ultimate method for continuously verifying and improving ontology quality, as well as for guiding its future development. Specifically, multiple issues can be found and fixed primarily through practical and diverse ontology reuse scenarios.

Abstract In this article we study and evaluate combinatorial algorithms for exploring state spaces of prototypical artificial intelligence problems that are represented by very large state space graphs. We are interested in the modeling and analysis of such graphs, including their visualization and quantification. We consider Blocks World as a prototype artificial intelligence problem that is often employed to introduce problem solving strategies using searching, planning, and reasoning algorithms. Our results are mathematical models and combinatorial algorithms supporting the concise definition of the Blocks World state space graph, as well as the exact evaluation of several metrics defined on this graph, including its number of states, its average number of stacks per state, its number of transitions, and its average branching factor. We also present experimental results supporting the effectiveness and efficiency of our proposed algorithms.

Abstract Explainable recommendations become essential when we need to improve the performance of recommendations and to increase user confidence. Explanations are effective when end users can build a complete and correct mental representation of the inferential process of a recommender system. This paper presents our view on the background regarding the implications of explainability applied to recommender systems. Our work contributes to the better understanding of the concept of explainable recommendation and it offers a broader picture of the development of further research in this field. Additionally, we contribute by providing a better understanding of the concept of human‐centered evaluation of explainable recommender systems.

In recent years, in the era of information overload development, the need for recommender systems that make personalized suggestion systems has become a very exciting field for researchers. To develop models that generate high-quality recommendations, the explainable recommendation has been introduced, proposing to develop intuitive and trustworthy explanations. The problem that the explainable recommendation wants to solve is to let people understand why certain elements rather than other are recommended by the system. This paper briefly overviews the short history of explainable AI and then it presents its role and applicability in the domain of recommender systems. Our work contributes to understanding the concept of explainable recommendation and what it should accomplish to increase its acceptability and to enable its accurate evaluation.

Blocks World is a prototype artificial intelligence problem used to exemplify problem solving using searching and planning algorithms. In this paper we present and experimentally evaluate combinatorial algorithms for measuring the size of the search graph associated to this problem. We focus on evaluating exactly a few metrics on the Blocks World state space graph, including the number of states, the average number of stacks per state, the number of transitions, and the average branching factor.

[No abstract available]

We have recently proposed a mathematical model of collective profitability, in semi-competitive intermediation networks. In this work, we are interested in determining optimal pricing strategies of network participants. The optimization criterion is defined using the Nash social welfare function. We provide theoretical results of existence of such strategies, as well as computational experimental results, based on nonlinear convex mathematical optimization.

The overarching connectivity of "things" in the Internet of Things presents an appealing environment for innovation and business ventures, but also brings a certain set of security challenges. Engineering secure Internet of Things systems requires addressing the peculiar circumstances under which they operate: constraints due to limited resources, high node churn, decentralized decision making, direct interfacing with end users etc. Thus, techniques and methodologies for building secure and robust Internet of Things systems should support these conditions. In this paper, we are presenting a description of the CAAVI-RICS framework, a novel security review methodology tightly coupled with distributed, Internet of Things and fog computing systems. With CAAVI-RICS we are exploring credibility, authentication, authorization, verification, and integrity (CAAVI) through explaining the rationale, influence, concerns and security solutions (RICS) that accompany them. Our contribution is a thorough systematic categorization and rationalization of security issues, covering the security landscape of Internet of Things/fog computing systems, as well as contributing to the discussion on the aspects of fog computing security and state-of-the-art solutions. Specifically, in this paper we explore the Authentication in Internet of Things systems through the RICS review methodology.

Abstract We propose a new framework based on Belief‐Desire‐Intention multi‐agent systems for the macroscopic modelling and simulation of continuous dynamic systems. The main idea is to break down the target system model into a collection of autonomous and loosely coupled interacting components endowed with clean message‐based interfaces and local intelligence. Each component is then mapped to a Belief‐Desire‐Intention agent that captures its state as a set of logical facts and its behavioural patterns as a set of plans. The system model can be described as a multi‐agent programme that is specified using the state‐of‐the‐art Jason agent‐oriented programming language. The approach is evaluated by considering a generalized graph‐based model of predator–prey systems. Our approach supports the configuration of the multi‐agent model with various differential equations integration methods, as required by the specific problem.

Bagging is a popular method used to increase the accuracy of classification, by training a set of classifiers on slightly different datasets and aggregating their output by voting. Usually, the majority voting is used for this purpose, or the plurality voting, when the problem has multiple class values. In this study, we analyze the influence of several voting methods on the performance of two classification algorithms used for datasets with different levels of difficulty. The results reveal that the single transferable vote can be a good alternative to plurality voting, although it has the drawback of a higher computational cost related to the calculation of preference ordering.

This paper introduces a simple generalized formal model for patrolling games that is suitable for decision making regarding the efficient use of scarce resources in surveillance applications. The model is then mapped to a multi-agent simulation using Jason agent oriented programming language. We present the details of the multi-agent model of the game, as well as experimental results that we obtained by simulating a sample patrolling game. The main contribution of the paper is the mapping of the formal game model, as inspired by game theory, to the more practical agent-oriented programming approach, as advocated by agent-oriented software engineering.

The structure and workflows of an organization can be modelled by business agents following a set of roles. The agents act according to the established rules, and can interact in order to achieve the organizational objectives. Business processes described by means of role-activity diagrams can be formalized and implemented as multiagent systems. In this paper we compare two implementations of business agents, in Jason and F#, and discuss the features of these programming languages with respect to the problem under analysis. We address business processes both from the perspective of individual agents and rules, and from the perspective of a generic knowledge-based business agent architecture.

Paxos is regarded as one of the most important protocols for distributed consensus in the presence of failures. It uses a number of 2f+1 processes to tolerate the benign failure of f processes, by splitting them into different roles with tailored tasks. The protocol presents multiple engineering challenges, two of which are pinpointing the cause for stalling and identifying the faulty processes. In this paper, we propose an alternative solution based on weighted argumentation frameworks by introducing Distributed Dispute Trees and a new role (the Skeptic) which ensures safety by attacking each condition that is not met in the Paxos protocol. The other processes must collaborate to defeat the Skeptic and achieve consensus.

Currently there is a lot of interest in developing applications to support the mobility and movement of resources and related features. One notable example is freight transportation management. Ontology-driven multi-agent systems are promising technologies for building systems that provide an intelligent service for freight transportation brokering aiming to create optimal transport policies such that transport costs are reduced to the minimum. We propose a freight transportation ontology and then we define a scenario of transport logistics to validate our proposed ontology. The evaluation is based on the careful examination of ontology concepts and properties that are required for capturing freight transportation requests and resources.

We are honored and pleased to open this special issue of the Concurrency and Computation: Practice and Experience journal focused on intelligent distributed computing. Intelligent distributed computing (IDC) faces the challenges of adapting and combining research results in the fields of intelligent computing and distributed computing. Intelligent computing develops methods and technology ranging from classical artificial intelligence, computational intelligence, and multi-agent systems to game theory. Distributed computing develops methods and technology to build systems that are composed of collaborating components deployed on computer networks. IDC sets the foundation for the development of the new generation of intelligent distributed systems. One direction of IDC is the application of nature-inspired computing methods such as swarm intelligence and genetic algorithms (GA) to develop better distributed systems. On the other hand, distributed systems can help to develop more efficient systems that employ the metaphor of bio-inspiration for solving computationally difficult problems. Another direction of IDC proposes the use of intelligent software agents for the development of open and scalable distributed systems that can engage in complex intelligent processes. Intelligent software agents are endowed with high-level properties including knowledge and reasoning that make them appealing for building intelligent distributed systems. This special issue brings to the reader new research results and applications of intelligent distributed computing, with a special focus on the synergies between intelligent software agents and bio-inspired technologies on one side and parallel and distributed computing on the other side. In particular, some of the papers address interesting applications of intelligent distributed approaches in the areas of trust, security, protein structure prediction, and Semantic Web Services (SWS). We believe that the papers selected for presentation in this special issue will be a valuable resource for researchers and practitioners working in the emergent field of IDC, with a special focus for those especially interested in the recent advancements of the aforementioned research areas that are part of IDC. This issue includes extended versions of selected papers from the Fourth International Symposium on Intelligent Distributed Computing (IDC 2010) and the collocated Second International Workshop on Multi-Agent Systems Technology and Semantics (MASTS 2010), both held on 16–18 September 2010, Tangier, Morocco. There were 61 papers from 20 countries submitted to IDC 2010 and MASTS 2010 from which only 32 were accepted and included in the proceedings. From the accepted papers, related to the topic of Intelligent distributed computing, seven papers with higher review scores were selected and invited to be extended and submitted to this special issue. Finally, after two peer-review rounds, all of them were carefully revised, extended, and improved, thus being judged acceptable for inclusion in this special issue. The first article 1, by Gabriel Ciobanu and Calin Juravle, is in the area of mobile software agents. In this article, the authors introduce a systematic approach for deriving the software architecture of a mobile agents' environment. It is remarkable that the authors' work has a solid foundation on formal specification grounded on process calculi. This is one of the few papers showing clearly that advanced software engineering technologies can go hand-in-hand with formal foundations for the rigorous development of real distributed systems. The second article 2, by Igor Kotenko, Alexey Konovalov, and Andrey Shorov, is in the area of agent-based simulation applied to computer security. In this article, the authors introduce an interesting approach for investigation of cooperative distributed attacks and defense mechanisms by combining agent-based modeling with computer network simulation. On the one hand agents proved suitable for convenient representation of botnets and defense components, whereas on the other hand specialized network simulation packages (such as OMNeT++ INET) were used for creation of the agents' network environment, as realistically as possible. On top of this synergetic architecture, the authors employed a high-level specification of botnet and defense scenarios supporting botnet propagation, management, and attacks, as well as botnet defense and legitimate network activities. The flexibility of the agent-based approach allowed the creation of experiments with realistic simulation scenarios comprising many agents playing different roles and possibly organized in various attack and defense teams. The third article 3, by Antonio González-Pardo, Pablo Varona, David Camacho, and Francisco De Borja Rodríguez Ortíz, is in the area of bio-inspired multi-agent systems. In this article, the authors introduce a new bio-inspired method to improve agent communication for collaborative problem solving using a network of agents. Their method combines (i) an optimal communication topology based on scale-free networks with (ii) a discrimination policy for the messages that are analyzed by an agent. The approach was experimentally investigated for solving the blind jigsaw puzzle problem. The discrimination policy is based on the identification of the sender agent identity; this means that only messages from recognized agents are considered, whereas the other messages are discarded. Several experiments were performed with different agent populations, by varying the communication network topology of the agents (using the probability of redirection of the network communication channels), as well as the memory size that defines the size of the local informational context of agents. The fourth article 4, by Michele Malgeri, Vincenza Carchiolo, Giuseppe Mangioni, and Alessandro Longheu, is in the area of computational trust. In this article, the authors introduce a distributed and secure algorithm based on TrustWebRank metric. The algorithm was experimentally assessed using a large and realistic data set extracted from Epinions recommendation network. The experimental results included the convergence time and the bandwidth complexity. They revealed a fast convergence and a significant amount of messages exchanged by peer nodes. The fifth article 5, by Cristina Bianca Pop, Viorica Rozina Chifu, Ioan Salomie, Mihaela Dinsoreanu, Tudor David, Vlad Acretoaie, Aliz Nagy, and Ciprian Oprisa, is in the area of SWS. In this article, the authors present two new bio-inspired methods for clustering SWS based on the metaphors of ants' and birds' intelligence that employ particle swarm optimization and ant colony optimization algorithms. The evaluation of the matching degree of two services is based on an original metric that is using a decomposition technique. The two methods were experimentally evaluated using the standard SAWSDL-TC service test collection. The performance evaluation metrics were the Dunn index and the authors' original Average Item-Cluster Similarity metric. The experimental results revealed that algorithms based on particle swarm optimization outperform algorithms based on ant colony optimization. The sixth article 6, by César Manuel Benitez, Rafael Parpinelli, and Heitor Silvério Lopes, is in the area of parallel bio-inspired computational methods. In this article, the authors present a parallel computational approach that investigates the hybridization of artificial bee colony (ABC) with GA. The resulted hybrid method is then applied to solve the protein structure prediction problem. The experiments were carried out on the Beowulf cluster parallel processing environment. The experimental results show that the hybrid ABC-GA approach outperforms the simple ABC approach. Finally, the seventh article 7, by Daniel Diaz, Salvador Abreu, and Philippe Codognet, is in the area of parallel constraint satisfaction. In this article, the authors introduce a parallel extension of constraint-based local search algorithm called adaptive search (AS). They experiment with a parallel version of AS called AS/Cell that was designed for the Cell Broadband Engine (Cell/BE). The experimental results show a linear speedup when scaling up the number of cores. Moreover, simultaneous exploration of the different parts of the search space sometimes leads to super-linear speedups. Finally, the randomness for the diversification of the search increases robustness of the results; that is, the differences between the minimum and maximum execution times, as well as the overall variance of the results, decrease significantly. Concluding, we would like to thank the authors of the papers for preparing extended versions of their conference papers and the reviewers for their great job that assures the high quality of the final articles. Also, we would like to thank Prof. Mohammad Essaaidi and Prof. Michele Malgeri for co-organizing IDC 2010, whose high scientific quality standard enabled this special issue. Finally, we would like to express our appreciation to Prof. Geoffrey Fox, Editor-in-Chief of Concurrency and Computation: Practice and Experience, for offering us the opportunity to edit this exciting special issue. We really hope that the readers of this issue will find the articles quite interesting and stimulating.

This paper analyzes the behavior of students in an educational hypermedia system, with the goal of identifying correlations between patterns of behavior and studentspsila learning style. Data from an exploratory study are presented and discussed: 22 students interact with an educational hypermedia system called WELSA, having all their actions monitored and logged by the system. Behavioral patterns are subsequently determined based on these actions and a subset of them are graphically sketched in this note.

Accommodating learning styles in adaptive educational systems represents an important step towards providing individualized instruction. The paper summarizes the main results reported in the literature, as well as the main criticisms of the current approaches. Thus, most of the adaptive educational systems to date are based on a single learning style model and the diagnosis is done explicitly, by means of a psychological questionnaire. The lack of an integrative, comprehensive learning style model and the questionable validity and reliability of the measuring instruments are two important problems of this approach. In this context, we advocate the use of a unified learning style model, which exhibits the following properties: i) it integrates the most relevant characteristics from several models in the literature; ii) it includes e-learning specific aspects (technology related preferences); iii) it is stored as a set of learning characteristics, not as a stereotyping model. Moreover, the student diagnosis is to be done implicitly, based on analyzing her/his interactions with the system.

Recently, we have proposed a comprehensive agent-based e-commerce system. While UML formalized, it lacked details how basic functions - e.g. user request to purchase a given product - are to be implemented. Furthermore, the "airline ticket reservation model" used in the system involves time management issues that have not been addressed. The aim of the paper is to discuss the way in which the information flow and data transformations involved in it are to be implemented; assuming that information about products is to be ontologically represented. Furthermore, a simple way in which time information can be successfully managed to support the proposed product reservation approach will be discussed

Many formal modelling notations for business processes have been proposed during the last decade. They can be broadly classified into high-level visual notations, with an intuitive meaning, mainly addressed to the business management community and low-level foundational notations, with a detailed and formal semantics, mainly addressed to the computer science community. Role activity diagrams are a popular high-level visual notation for capturing the dynamics and role structure of an organization. This paper establishes that role activity diagrams have a formal semantics as well and thus making them suitable to formal verification. The result is obtained by mapping of a role activity diagram model to a process algebra model. Process algebras are mathematical languages for the specification and understanding of concurrent and cooperating computational processes.