Louise J. Seixas

A multi-agent intelligent environment for medical knowledge

AMPLIA is a multi-agent intelligent learning environment designed to support training of diagnostic reasoning and modelling of domains with complex and uncertain knowledge. AMPLIA focuses on the medical area. It is a system that deals with uncertainty under the Bayesian network approach, where learner-modelling tasks will consist of creating a Bayesian network for a problem the system will present. The construction of a network involves qualitative and quantitative aspects. The qualitative part concerns the network topology, that is, causal relations among the domain variables. After it is ready, the quantitative part is specified. It is composed of the distribution of conditional probability of the variables represented. A negotiation process (managed by an intelligent MediatorAgent) will treat the differences of topology and probability distribution between the model the learner built and the one built-in in the system. That negotiation process occurs between the agents that represent the expert knowledge domain (DomainAgent) and the agent that represents the learner knowledge (LearnerAgent).

ESTRATÉGIAS PEDAGÓGICAS NO ENSINO DE ALGORITMOS E PROGRAMAÇÃO ASSOCIADAS AO USO DE JOGOS EDUCACIONAIS

It is difficult to work in classroom with the introductory content of algorithms and programming, and it poses many problems that make the students give up. The reasons for this include the lack of motivation of the students and their difficulty in developing the necessary logical reasoning for the construction of algorithms. This paper proposes the use of pedagogical strategies, such as computational games, to mitigate these problems, suggesting that the use of computational resources in the area of computer education might be interesting.

A model of pedagogical negotiation

This paper presents a model of pedagogical negotiation developed for the AMPLIA, an Intelligent Probabilistic Multi-agent Learning Environment. Three intelligent software agents: Domain Agent, Learner Agent and Mediator Agent were developed using Bayesian Networks and Influence Diagrams. The goal of the negotiation model is to increase, as much as possible: (a) the performance of the model the students build; (b) the confidence that teachers and tutors have in the students’ ability to diagnose cases; and the students’ confidence on their own ability to diagnose cases; and (c) the students’ confidence on their own ability to diagnose diseases.

Formal analysis of a probabilistic knowledge communication framework

This paper introduces a new formal model, which generalizes current agent communication theories (basically the FIPA version of these theories) to handle probabilistic knowledge communication. Several questions about communication of probabilistic knowledge are discussed in the light of current theories of agent communication and it is argued that exists a semantic gap between these theories and research areas related to probabilistic knowledge representation and communication. This gap creates serious theoretical problems if agents that reason probabilistically try to use communication framework provided by these theories. To diminish this gap it is proposed a modal probabilistic logic and a new communication framework composed of communication principles and acts for probabilistic knowledge communication.

Formal Analysis of the Communication of Probabilistic Knowledge

This paper discusses questions about communication of probabilistic knowledge in the light of current theories of agent communication. It will argue that there is a semantic gap between these theories and research areas related to probabilistic knowledge representation and communication, that creates very serious theoretical problems if agents that reason probabilistically try to use the communication framework provided by these theories. The paper proposes a new formal model, which generalizes current agent communication theories (at least the standard FIPA version of these theories) to handle probabilistic knowledge communication. We propose a new probabilistic logic as the basis for the model and new communication principles and communicative acts to support this kind of communication.

Estratégias Pedagógicas para um Ambiente Multi-Agente Probabilistico Inteligente de Aprendizagem - AMPLIA

Este trabalho pretende avaliar se e possivel elaborar estrategias pedagogicas com base em modelos de niveis de tomada de consciencia e utiliza-las, por meio de agentes inteligentes, em um ambiente de aprendizagem. O ambiente utilizado foi o AMPLIA - Ambiente Multi-agente Probabilistico Inteligente de Aprendizagem, desenvolvido inicialmente como um recurso auxiliar para a educacao medica: neste ambiente, o aluno constroi uma representacao grafica de sua hipotese diagnostica, por meio de uma rede bayesiana. O AMPLIA e formado por tres agentes inteligentes, o primeiro e o Agente de Dominio, responsavel pela avaliacao da rede bayesiana do aluno. Os projetos dfos outros dois agentes inteligentes do amplia AMPLIA sao apresentados nesta tese: o Agente Aprendiz, que faz inferencias probabiisticas sobre as acoes do aluno, a fim de construir um modelo do aluno baseado em seu nivel de tomada de consciencia, e o Agente Mediador, modelo do aluno baseado em seu nivel de tomada de consciencia, e o Agente Mediador que utiliza um Diagrama de influencia para selecionar a estrategia pedagogica com maior probabilidade de utilidade. Por meio de uma revisao dos estudos de Piaget sobre a equilibracao dasestruturas cognitivas e sobre a tomada de consciencia, foi construida a base teorica para a definicao e organizacao das estrategias. Essas foram organizadas em classes, de acordo com o principal problema detectado na rede do aluno e com a confianca declarada pelo aluno, e em taticas, de acordo com o nivel de tomada de consciencia inferido pelo Agente Aprendiz. Foram realizados experimentos praticos acompanhados por isntrumentos de avaliacao e por observacoes virtuais on line, com o objetivo de detectar variacoes nos estados de confianca de autonomia e de competencia. Tambem foram pesquisados indicios de estados de desequilibracao e de condutas de regulacao e equilibracao durante os ciclos de interacao do aluno com o AMPLIA. Os resultados obtidos permitiram concluir que ha evidencias de que, ao longo do processo, ha ciclos em que o aluno realiza acoes sem uma tomada de consciencia. Estes estados sao identificados, probabilisticamente, pelo agente inteligente, que entao seciona uma estrategia mais voltada para um feedback negativo, isto e, uma correcao. Quando o agente infere uma mudanca neste estado, seleciona outra estrategia com amior utilidade para dar inicio a um processo de negociacao pedagogica, isto e, uma tentativa de maximizar a confianca do aluno em si mesmo e no AMPLIA, assim como maximizar a confianca do AMPLIA no aluno. Os trabalhos futuros apontam para a ampliacao do modelo do aluno, por meio da incorporacao de um maior numero de variaves, e para a necessidade de aprofundamento dos estudos sobre a declaracao de confianca do ponto de vista psicologico. As principais contribuicoes relatadas sao na definicao e construcao de um modelo de aluno, com utilizacao de redes bayesianas, no projeto de um agente pedagogico como mediador num processo de negociacao pedagogica, e na definicao e selecao de estrategia pedagogica para o AMPLIA.