Virtual.Prob - Ambiente virtual para auxÃlio e verificaÃÃo do aprendizado

AUTOR(ES)

MÃrcia Hafele IslabÃo Franco

DATA DE PUBLICAÇÃO

2004

RESUMO

This work presents a virtual learning environment, Virtual.Prob, whose objective is to help the student to study Probability in the discipline Statistics and Probability for Computing (ET 586). This discipline is part of the Computer Engineering and Computer Science courses at UFPE. Virtual.Prob is based on the machine learning technique known as Bayesian Networks (BN). The BN are formed by a qualitative part and a quantitative one. The qualitative part is the Graphical model (Directed Aciclic Graph -DAG) where the variables are the nodes that compose the net. This also includes the rules (links), that represent dependence relations between the variables. The quantitative part is the set of conditional probabilities associated to the existing arcs in the graphical model and the estimated probabilities of the variables.In Virtual.Prob each student has a BN. The probabilities that represent the performance of the student in relation to the content of each node are stored in his BN.The environment has also a virtual assistant who gives suggestions to the user in accordance to his performance. In Virtual.Prob, besides the contents related to Probability (which include examples and exercises), other resources are also available, namely: help, notice board, and a Web browser. Virtual.Prob has a monitoring system for the performance of each student. The professor or monitor has access to this system, where the supplied information became an important tool to verify which specific topics the students are finding difficulties. The student can access his own performance, so that he can verify in which topics he should concentrate efforts. The Virtual.Prob environment was tested, with success, by students of the courses of Computer Science and Computer Engineering (CIn-UFPE). The results of our experiment include suggestions of further work, such as the expansion of the BN and the knowledge base, refinement of the virtual assistant, implementation of a corrector of answers, graphical visualization and a chat room

ASSUNTO(S)

redes bayesianas learning environments probability probabilidade bayesians networks ciencia da computacao ambientes de aprendizagem

Documentos Relacionados

• Ambiente virtual distribuÃdo - auxÃlio ao desenvolvimento de tÃticas da aviaÃÃo de combate.
• Desenvolvimento organizacional com auxÃlio do desdobramento da funÃÃo qualidade.
• Sistema de simulaÃÃo sonora para auxÃlio em treinamento em cardiologia.
• Ensinus: a general purpose virtual environment for verification and has supported to the learning
• DiscretizaÃÃo para aprendizagem bayesiana: aplicaÃÃo no auxÃlio à validaÃÃo de dados em proteÃÃo ao vÃo.