This study investigates the effectiveness and feasibility of using parallel machines with GPUs of different capacities to train large language models (LLMs) as an alternative to costly cloud platforms. The study explores optimization techniques focused on data, models, budget, and systems, detailing the configuration of a multi-GPU architecture and its implementation with CUDA software, PyTorch, and the HetSeq library, adapted to maximize performance in heterogeneous systems, including the usage of legacy resources (e.g., older GPUs). In performance tests, three approaches are compared: homogeneous load distribution, heterogeneous distribution, and HetSeq usage. The HetSeq configuration yielded a substantial performance improvement, achieving an execution time of 129,759 seconds, significantly lower than the 153,175 seconds observed for the homogeneous configuration and 165,387 seconds for the heterogeneous configuration in the maximum setup tested, representing a reduction of 15% and 21%, respectively. These results highlight HetSeq's advantage in optimizing training time as more GPUs are added, outperforming traditional approaches that attempt to standardize hardware utilization, such as PyTorch. Further analysis using Amdahl's Law reveals that approximately 82% of the training process is parallelizable, underscoring that optimized heterogeneous architectures are both viable and economically advantageous in scenarios that do not require homogeneous hardware.

The integration of AI with IoT heralds the era of AIoT (Artificial Intelligence of Things). It represents a transformative approach in technology and opens up a new opportunity for deploying machine learning models in embed-ded devices that face resource constraints and operate on the edge of networks. Central to this study is the implementation of computational vision techniques for digit recognition, evaluating various machine learning models, particularly in the context of smart metering. The selected models were converted from GPU-equipped workstations to ESP32-S3 microcontroller-based low-end devices. Through a series of experiments using ESP32-S3 development kits, the MNIST database, and TensorFlow Lite, we explore the effectiveness of these models in smart metering applications, focusing on accuracy, inference times, and the challenges in model conversion. The findings demonstrate the feasibility of executing machine learning inferences on low-end devices with high accuracy in smart meter contexts. However, challenges such as model size limitations, processing speed, conversion difficulties, and potential accuracy loss were noted. Not all models were viable for conversion to TensorFlow Lite. Simpler models like LeNet5 emerged as effective solutions for smart metering applications, balancing size, accuracy, and latency. This work offers practical insights for researchers and engineers looking to implement machine learning in AIoT and smart metering environments, highlighting the trade-offs and considerations for effective deployment.

The virtualization of natural environments is becoming increasingly common in contemporary society, and it has found even more widespread application due to social restrictions, practical considerations, and economic factors. This study presents a dynamic generator of immersive collaborative environments in CAVE (Cave Automatic Virtual Environment), which employs stereoscopy to project 3D images onto a simulated CAVE. The environment supports real-time interaction with the projected elements, and the outcome of the project is a system that can dynamically generate virtual environments in CAVE of varying sizes and formats, rendering elements that can be projected onto a real model.

Process Mining (PM) is a research discipline that helps organizations track and optimize processes to support their business. Further, it focuses on providing process analysis techniques and tools, and several of its applications have been described in the literature. The start point for PM is using event logs generated by information systems to analyze processes. These event logs need to be extracted from databases and prepared for use because the quality of the event logs used as input is critical to the success of any PM effort. In this article, we present a systematic mapping review to provide the reader with highlights of the state-of-the-art techniques for event log preparation. Based on the retrieved studies, we identified six main categories of log preparation techniques: extraction, cleaning, repair, non-adequate granularity, quality evaluation, and privacy. The results are explored quantitatively and qualitatively. All results are made available through spreadsheets and charts. We believe this paper is a starting point for researchers to identify the studies that would help them prepare event logs for PM.

Process Mining (PM) is a research area that focuses on helping organizations optimize their business processes. PM provides process analysis tools and techniques with a wide range of applications described in the literature. Additionally, PM can be associated with Root Cause Analysis (RCA), a problem-solving technique based on the assumption that a problem can only be solved by addressing its underlying cause. In this study, we performed a case on a real-world Brazilian telecommunication company that provides telephony, television, and broadband Internet subscription services. We applied PM and RCA to identify the possible causes for short and long-duration services, high rework rates, and activity repetition. We split the dataset into three aspects: Support, Cancellation, and Sales related services and analyzed them separately. For RCA, we applied both decision tree classification and rule mining. We identified 53 rules over 9 decision trees that were validated by process specialists from the tech company.

Surgical center scheduling is challenging to schedule physical spaces to reduce costs and increase productivity. This study presents a framework called PM4SOS (Process Mining for Simulation, optimization, and Scheduling) that facilitates the generation of an operating room schedule in an automated way and integrates, reducing cognitive overload and waste of time. This framework allows the evaluation of restrictions to get the best performance of surgical schedules. PM4SOS combines process mining with data from event logs, generation of the automated simulation model, and case-based reasoning (CBR) to analyze management indicators, allowing significant gains in optimization of the scheduling of surgical centers. The results with PM4SOS help decision-making in hospital environments to better use physical resources and human resources, such as reducing waiting time, optimizing surgery execution time, and resource capacity management.

When used together, process mining and simulation provide better indicators for informing business management. The generation of simulation models involves significant effort, time, and system integration. The combination of process mining techniques and short-term simulation requires the extraction of real-time data from organizational systems. However, from the perspective of short-term effects and their reduced time, extraction and simulation tasks require automated execution. This study presents a framework called PM2Sim, which allows users to generate a simulation model in an automated and integrated manner with reduced effort. This framework was built on an open-source environment and event logs were used as a data source. The results obtained with PM2Sim support operational decision making and visualize scenarios of control, such as management of waiting time, process execution time, and resource capacity management.

The growth in use of Blockchain technology has enabled the development of applications in various industries. In particular, there has been an interest about using it in government to manage electronic public services, such as configuring and improving electronic government (e-government). However, the in-depth study of practical Blockchain solutions in the government sector is still under-researched despite the growing academic interest in this area. Therefore, this article presents a systematic literature review based on the current technology relating to Blockchain architectures applied by governments to public services.

Organizations invest many resources and time for improving business process and cooperative work. Traditional process mapping requires a lot of effort to diagnostic performance issues and to understand the main causes of losses. Process mining emerges as a new discipline focused on analyzing process based on real event data aimed to automate discovery of process models; to check conformance and to extend models performance or resource analysis. This paper combines a discovery process mining and a process variant clustering algorithm, focused on obtaining knowledge for a top-down navigation concerning performance cause analysis. An applied industry case was conducted to verify the proposed techniques using a dataset extracted from an ERP. From the results obtained, it was possible to identify the main performance losses segregated by process variants.

This article presents a multilevel model to support decision-making in accepting or not a recommendation. Acceptance of a recommendation occurs in an open and uncertain environment. To deal with this context, the proposal is to take into account the judgments - or the preferences of the decision maker - to derive degrees of performance and risk; this allows changing the context of decision making, from pure uncertainty to measurable risk. The proposal also considers, with less emphasis, information on the trust and reputation of a referral agent. The evaluation scenario describes a relationship between a contracting agent and a set of service providers; the best-reputed provider and lowest-risk proposition is selected. The result is an approach that allows you to configure the behavior of a contracting agent to act in a more risk-prone or restrictive manner.

The sales and technical engineering processes are neither sequential nor deterministic. Usually there are iterative feedback loops and optional steps during many activities like contract negotiations, product modification, specialist consulting, design calculations, management approval, validation tests or certification process. In large and complex manufacturing processes, the sales and product customization process is a challenge and significant efforts are required to understand and improve it. This paper presents an industry process mining application as a way to reduce efforts in the diagnostic and control phases of an improvement process project. The process mining was based on a multifaceted method to identify the most significant issues in control-flow, organizational structure, and performance. The approach used to discover the processes was the Fuzzy Miner algorithm, because it is robust in dealing with noise in the logs and unstructured processes. As a result, there was a reduction of efforts in the mapping phases that were much faster than the traditional process mapping. The work reported concerns events at a company related to the sales order up to the production process that are characterized as less structured.

Trust mechanisms are considered the logical protection of software systems, preventing malicious people from taking advantage or cheating others. Although these concepts are widely used, most applications in this field do not consider affective aspects to aid in trust computation. Researchers of Psychology, Neurology, Anthropology, and Computer Science argue that affective aspects are essential to human's decision-making processes. So far, there is a lack of understanding about how these aspects impact user's trust, particularly when they are inserted in an evaluation system. In this paper, we propose a trust model that accounts for personality using three personality models: Big Five, Needs, and Values. We tested our approach by extracting personality aspects from texts provided by two online human-fed evaluation systems and correlating them to reputation values. The empirical experiments show statistically significant better results in comparison to non-personality-wise approaches.

This paper presents a trust model that aggregates multiple sources of information, such as: testimonies, context, social information, preconception, reputation, and proposes a new data structure, called Dossier. The aim is to raise the success rate that consumer agents have when contracting provider agents. To reach this objective we make two hypotheses: (i) every consumer agent is able to measure and inform the quality of a service they receive from a provider agent; (ii) the Dossier is a data structure that allows the evaluated agent himself to keep locally the history of feedbacks received without modification or omission that benefits him. The model-that uses the Dossier structure-reduces some weak points of trust models that rely on the direct interaction between provider and consumer agents (direct trust) or those that rely on testimony obtained from client agents (indirect trust). The Dossier model allows us effectively to use the agent approach to build decentralized IT solutions, where control and data-of every application encapsulated in an agent-can actually be distributed logically and physically. The comparative analysis showed that the Dossier model has a behavior equivalent to or superior to the considered models (FIRE, TRAVOS, MARSH and Central) and that such a model can make the agents more efficient in choosing their partners. The reason for this is that the reputation of a given service provider agent is based on the reputation it has among the totality of consumer agents that used its services.

This paper presents a model for supporting the decision-making of a contracting agent, in the light of the assignment of a contract to a service agent. Such contract, as an operational instrument, describes the causes of the contracting agreement. The performance of the contractor takes place in an open and uncertain environment. To deal with such context, we consider judgements that allow deriving degrees of risk in addition to trust and reputation information about a service provider agent. The evaluation scenario describes a service contracting relationship between a contracting agent and a set of potential providers. Results show that the model involving risk calculation is more efficient when seeking for a conservative behavior for the system.

The reduction in cost, sequencing time and assembly of thousands of complete genomes imposed an increasing demand for genomic data storage. Data compression approaches are viable solutions to circumvent such a demand and can reduce genetic data by more than 75% of the needed space amount of uncompressed information. Current specialized strategies implemented as compression tools for DNA sequences take advantage of the inherent characteristics of the DNA sequences, such as the small size of the required alphabet (ATCG) for data representation as well as the frequency of internal repeats in a sequence. In this work, we explore a novel perspective for lossless genomic sequence compression using standard lossless image formats. We show that DNA sequences stored in a FASTA file are converted into a matrix of bits. The matrix was then encoded in the lossless image file formats WEBP and FLIF for performance evaluation. It was performed using a dataset containing 1,547 DNA sequences from 1,163 different organisms from the kingdoms Animalia, Archaea, Bacteria, Fungi, Plant, and Protist. Our results demonstrated that the WEBP format presented the best space savings (76.15%, ±0.84) when compared to the other image formats. Furthermore, when compared to specialized genomic data compression tools, WEBP space savings presented similar results with DELIMINATE (76.03%, ±2.54) and MFCOMPRESS (76.97%, ±1.36).

This paper presents a system applied in the automatic driving train using Case Based Reasoning (CBR) with Differential Evolution (DE). CBR was used to retrieve, reuse and revise experiences from real data during the journey. The DE was used to adapt the cases retrieved and optimize then considering multi-objective optimization. For this purpose, a train driving simulator used and the results compared the data of real train driving scenarios. Multi-objective optimization was used to reduce fuel consumption and also travel time. The results obtained concerning fuel consumption was entirely satisfactory because in some cases average savings of 45% in fuel consumption about the results obtained by human drivers. Adapting cases using the Differential Evolution approach led to a 5% gain in consumption over an adaptation of cases using Genetic Algorithm. It should also note that the time required for case adaptation was lower using Differential Evolution Algorithm than using Genetic Algorithm.

Process mining allows observing process execution based on real event data and proposes methods and tools to provide diagnostics, reducing the gaps between practice and conceptual models. When process mining discovery techniques are applied in flexible processes with many decisions at runtime, the results are often semi-structured or unstructured process models that are difficult to understand. In this context, trace clustering is an approach for reducing the complexity of process models and improving the accuracy and comprehensibility. This paper presents an applied case in industrial manufacturing production with an unstructured process and issues in production performance indicators. A set of techniques were used to understand how the process occurs in practice, how many trace clusters should be identified as homogeneous process variants, and what causes production inefficiency. Finally, the results identify bottlenecks caused by erroneous decisions at runtime and serve to support process improvement.

This paper demonstrates the use of a classical Case-Based Reasoning (CBR) approach applied to the automatic train conduction scenario. We use a CBR model, where the adaptation task consists on a multi-objective optimization approach. To realize the case study we have used a train simulator. It is capable of conducting a train in a pre-defined railway providing relevant data about the conduction, such as, travel time and fuel consumption. In the experiments, we compare the data from the case study with a real scenario. The results are very promising considering fuel consumption and travel time; in some scenarios the economy of fuel was about 51%, and, we obtained a reduction about 69.5% of travel time in the best-case scenario. The CBR architecture proved to be efficient to the case study, because the past experiences assured a more efficient and economic train conduction.

The problem of service selection and composition is central to the field of Semantic Web Services (SWS). Developing effective selection and composition of services can not only produce dynamic service-based environments, but also help business teams and information system developers to better satisfy their user requirements. We propose a new framework, the objective of which is to go one step further during the business process modeling activity and to help analysts design their models, expressing some steps of a business process in terms of goals instead of directly attaching SOA services. Our work proposes an approach to select and combine services using Intelligent agents and planning techniques, using a goal as a source. With this approach we can produce a solution containing one or more services to reach the given goal. We propose MOSS as a formal language to describe the preconditions and effects of services and the goals of agents, and show how the evaluation mechanism can be used to check the preconditions in a given representation of the world, build the potential effects before the execution, and determine if a service will contribute to the goal of an agent.

This article presents a multi-phase bidding model that is primarily for reverse bidding (1:N) and can also be used later for bilateral bidding (1:1). This multi-phase approach excels by competing for the lowest price and relativizes a subtle “lose-win” relationship of its own for the reverse auction through a second phase of negotiation. The latter is limited to a bilateral relationship and is applied, if necessary, between the purchaser and the second or third best offer of the reverse auction. Experiments were conducted with stationary or adaptive negotiation agents using learning techniques to conduct negotiation policy, using a genetic algorithm to characterize the opponent's preferences and configure the generation of interesting offers. The results showed the influence that an aggressive bidder has on the process as a whole and also what can be done to minimize this effect.

This paper presents a planning approach using Case-Based Reasoning (CBR) modeled as a Subsumption Architecture to generate plans for driving trains. The main idea of a planning strategy is to generate a sequence of actions for an agent, which can use these actions to change its environment. CBR allows using prior experiences for new task assignments. In the proposed ap-proach, each previous experience (if not applicable) is adjusted us-ing one or more adaptation methods like substitutive and genetic algorithm. Our interest is to create a flexible architecture for an agent and apply it to simulate train conductions. We expect that the plans generated by this approach generate better results com-pared to another studies already developed for the area mainly considering fuel consumption and travel time.

Service-oriented architecture (SOA) is a proven approach that aims at producing loosely coupled, standard-based, and protocol-independent services. A compliant SOA architecture must provide independents units or services, allowing users to discover, execute and compose them in their applications. In order to follow the rapidly changing and highly competitive market, organizations have to adapt their service interfaces according to their business requirement. Thus, the design is one of the most crucial phases of service lifecycle. From the last years, the software engineering domain has made a great effort providing methods for agile development. Despite these efforts, we lack the necessary tools to use and validate the rich vocabulary presented in user stories and test cases during the service design. We propose a CSCWD approach for the exploitation of independent services using a personal assistant to guide the end user through a natural language dialog. The personal assistant plays the role of a mediator between end-users and the service library. The generated proof-of-concept allows the interaction with services through a personal assistant using restricted requests in natural language. Using our approach, software designers and domain experts can evaluate the expressiveness of their service interfaces and conduct a process improvement, adjusting the vocabulary and granularity of services iteratively.

This paper presents an Intelligent System, based on a dynamic time table definition, which coordinates the overtaking process of trains traveling in the same section of a railroad through a crossing loop. Each train involved on the overtaking process is represented by an intelligent agent capable of taken his actions based on his relative position on the railroad and his scheduling. The main goal of these agents is to react during the driving to avoid that more than one train stays on the same section of track at the same time (resource concurrency), and also to avoid unnecessary halts. The agents actions are previously defined, based on a simulation of the journey to the next crossing loop. For each stretch the agents selects another agent to be his coordinator, based on specific criteria. Then, each agent creates its action policy based on his local view and the data of the coordinator. The coordinator receives the actions and validates them generating a time table containing the actions for the next stretch, called dynamic time table. The definition of the action policy occurs dynamically, as each agent simulates the next step of the journey virtually and takes the decisions at runtime. The communication between the agents is done through the environment, where each agent updates his relative location and time. The main goal of the Intelligent System is the coordination of the trains focusing on reducing fuel consumption and also reducing the travel time. This is chased avoiding unnecessary halts, collisions of the trains and driving the trains with a Cruising Speed. The best simulations results achieved a 33.72% reduction in fuel consumption and a 33.30% reduction on travel time.

In this paper we propose an architecture of intelligent agent for automatic locomotives operating. The system agent generates its action policy using a set of resources, such as type of railway, composition, belief perception and reasoning about the actions. The focus of the operator agent is directed to the choice of acceleration points (gear) and preparation of travel plans in a journey guided by goals and objectives. The system is equipped with a module capable to plan the actions to move the vehicle from an initial point P to an end point Q and an executor module that implements the generated plan and modifies the state of the environment. For this purpose, we use the mental model that is based on the triple Belief, Desire and Intention (BDI) to which the perception of the agent is guaranteed by a set of sensors that provide speed information, position and breaks condition. The main focus on this research is the usage of mental model BDI for the resolution of a problem that combines travel naturally conflicting factors, such as safety, time and fuel consumption. Experimental results show that the developed architecture using the mental model BDI increases the efficiency of autonomous vehicles operating.

This paper presents a planning approach using Case-Based Reasoning (CBR) to generate plans for driving trains. The main idea of a planning strategy is to generate a sequence of actions for an agent, which can use these actions to change its environment. CBR allows using prior experiences in the situation assessment task. In the proposed approach, each previous experience (if not applicable) is adjusted resulting in cases specializations. Our interest is reducing the number of corrections triggered when a case retrieved is not applicable, based on these specializations. Experiments showed that the plans generated using this proposed method had a significant increase in the number of cases recovered satisfactorily, also reducing the need of adaptations for the cases recovered.

This paper presents a methodology to obtain rules of conduction from a set of data captured from sensors placed at a train as well data of actions executed by drivers. These actions result in a history H. The knowledge discovered is put in practice in a driving simulator and the result of the simulated actions generates a history H'. The validation of the discovered knowledge is done in an objective manner, which is calculated as a degree of similarity between the records. This degree of similarity reflects the performance of knowledge discovery process, which in experiments was around 85%. This degree of similarity represents how next were H and H.

This paper describes an intelligent approach based on agents that are able to drive and coordinate trains on stretches of railway line containing a crossing loop. Halts close to or even in crossing loops lead to increased consumption of fossil fuels, longer journey times and exhaustion of track capacity. In this paper the agents make use of a set of resources - railway line characteristics, train characteristics, driving rules and information about other trains - to generate their action policy. The agents perception is guaranteed by a set of sensors that provide data such as speed, position and information about the line. The tasks that the agent performs include carrying out actions such as increasing or reducing the speed of the train. The main objective of this study was to avoid unnecessary halts, which are the main cause of increased fuel consumption and journey time. Our results show that strong reductions can be made in terms of fuel consumption (average reduction of 25.5%), journey time (average reduction of 22.5%) and exhaustion of track capacity. Simulations were performed in which traditional driving techniques, with halts at several points along the stretch of track, were compared with driving performed by the multi-agent system, without any halts.

This paper presents an intelligent approach based on software agents capable of conducting and coordination trains in stretches of single railway track, aiming to reduce the utilization of railway and environment impacts. In the Brazilian rail modal, due to the low duplication of tracks, trains that journey on single railways should accomplish required halts, in order to wait for other trains to use the crossing loop safely. The technological evolution resulted on the appearance of new railway traffic system control. However, systems that rely on software agents are not well explored yet. Therefore, this paper elaborated a Multi-Agent System capable of simulating railway environment using agent drivers and agents with a highest level in managing the railway tracks. The behaviour of agents was based on specialized rules of conduction. The coordination between them occurs through message exchanges, always aiming to avoid halts during the journey. Results have shown an strong average reduction of 22.5% in journey time and 25.5% in fuel consumption when compared to journeys using the traditional method of conduction. The reduction, not only on the journey time, but also on the fuel consumption, entails on the decrease of CO2emission.

This paper presents the development of an intelligent agent used to assist vehicle drivers. The agent has a set of resources to generate its action policy: road and vehicle features and a knowledge base containing conduct rules. The perception of the agent is ensured by a set of sensors, which provide the agent with data such as speed, position and conditions of the brakes. The main agent behaviour is to carry out action plans involving: increase, maintain or reduce speed. The main effort of this research was the induction of conduct rules from data of previous trips. These rules form a classifier used for the selection of actions forming the conduction plan. Results observed with the experiments have showed that the proposed classifier increases the efficiency throughout the conduction of vehicles.

We present a model for the certification of trust in multi-agent systems based on encryption. The objective is to raise the level of efficiency that client agents have when contracting specialized service agents. We make three hypotheses: (i) client agents are able to measure and inform the quality of a service they receive from a service agent; (ii) distributed certificate control is possible because every service agent stores the certificates it receives from its client agents and, (iii) the content of a certificate can be considered safe as long as the public and private keys used to encrypt the certificate remain safe. This approach reduces some weak points of trust models that rely on the direct interaction between service and client agents (direct trust) or those that rely on testimony obtained from client agents (propagated trust). Simulation showed that encrypted certificates of trust improved the efficiency of client agents when choosing their service provider agents. The reason seems to be that the reputation of a given service provider agent is based on the reputation it has among the totality of client agents that used its services.

This paper introduces a new agent-based algorithm for scheduling in CSCWD. Distributed artificial intelligence provides a lot of research areas, including CSCWD, with efficient decentralized optimization and problem solving techniques. In this paper we focus on how DCOP (distributed constraint optimization problem) can be used for scheduling in CSCWD, discussing a new algorithm. Our algorithm is almost-complete, providing the best solution most the times. However, it saves a lot of computational resources, being useful in situations where other state of the art algorithms are not feasible. The results are quite encouraging, showing that our algorithm outperforms easily two well-known DCOP algorithms in terms of runtime, number of messages, size of messages and throughput.

This paper proposes an approach to generate offers and counter-offers in a bilateral negotiation process between cognitive agents with the learning machine capacities. The approach is configured, as each participant of a negotiation process improve their satisfaction degree, gaining knowledge from prior experience, i.e., in the next negotiation session, each participant can individually use this knowledge to redefine the configuration parameters of strategies and tactics to generate offers and counter-offers. Each agent refines their strategies using a genetic algorithm application based on the historical offers and counter-offer exchanges dataset. This context can lead to a new dimension in CSCW system development. This approach was tested in a simulated bilateral negotiation environment, which involved, for example, a buyer agent and a seller agent. The discussions about the results confront different negotiation sessions comparing agents provided with strategy and tactics reconfiguration capabilities and agents using static strategies and tactics.

A common problem in organizations is to identify people with the right competencies to form a specialized team, in academic or industrial environments, which is capable of executing a self-managed project of Research and Development (R&D). This work presents a technique that allows for identifying people who have the most appropriated competencies to form a R&D team extracting information from their curriculum vitas (CVs). Information extraction in this work is carried out by means of textual retrieval techniques in document databases. The system was evaluated with data of real projects producing expressive results when identifying people to participate in research projects.

This paper aims to conceive a system capable of assessing the understanding an individual possesses on an arbitrary subject. A hard, essential question regarding a collaborative environment is how to evaluate the knowledge an individual maintains about a specific domain. Such evaluation is usually accomplished with collective works or examinations applied individually. However, techniques such as those are subjected to idiosyncratic factors related to both the tutor (level of experience, mood, affinity, etc.) and the team member. In this paper, we used some techniques from the field of natural language understanding based on concepts like dynamic memory, case-based reasoning and semantic parser. We discuss experiments, carried out within one such educational environment, which turned out to be coherent according to the opinion of the tutors involved

The decision support systems are applied where the problem is complex and the necessary information for the best decision is difficult of being gotten and being used. Amongst the used elements to assist the decision maker can detach the databases as a collection of facts and information and the interface with the users who assist the decision makers to interact with the systems. We consider the construction of contextualized ontology using a methodology that tries to not only facilitate the understanding of questions elaborated in natural language through context and for search of terms, as well as elaborating a reply in the same context of the question. It allows, of this form that a user elaborates research and refines or directs questions to the measure that goes receiving the answers to the questions.

Changes in work and its practice are leading corporations to valorize workers not only for technical competencies but also social ones. This tendency has consequences both for workers who need to improve such competencies (most of the time on the fly), and for students who have to be prepared for such market. The aim of this work is to show how concepts from social learning theory and communities of practice are being used in the design of a learning environment that uses collaborative problem based learning (PBL) and portfolios. The environment encourages social interactions among its users where the development of social competencies is concerned. The work is divided in three parts; the first describes the collaborative learning model, the second describes how implementation evolved and finally considerations of what has been done and what future work is likely to be.

Describes an approach to an intelligent tutorial system called CARDIOLOG, specially developed to help in teaching electrocardiogram (ECG) interpretation to medical students. Although the system can be used as a self-learning tool, one of its main characteristics is its adequacy for a classroom environment t contains a professor module and many student modules. Medical students and professors, physicians and people interested in the subject are able to form a solid knowledge base or complement their studies with CARDIOLOG. The activities rely on user-friendly interfaces, based on a question/answer model, with the possibility of consulting reference information (if enabled by the professor). Analyses of students' performance, reports and knowledge base maintenance tools are also available in CARDIOLOG. CARDIOLOG was implemented using artificial intelligence technologies and developed using object methodology, a client/server structure and a 32-bit Windows platform.