Ángel Neira

An Architecture for on-Line Intelligent Control of Blast Furnaces

This paper describes an expert system for the on-line diagnosis of blast-furnaces. The system monitors and takes the decisions through the analysis and evaluation of more than a hundred parameters considered as input variables. The main difference between the system proposed here and a classical diagnostic system is the coexistence of three different models of behaviour: one based on a theoretical model of behaviour of thermal status and permeability, a second empirical model based on the considerations given by the human experts, and a third model derived from the study of the real behaviour observed in the furnace over time, obtained by means of the study of historical files, using machine learning techniques.

An Expert System for Ironmaking

This paper describes an expert system for the on-line diagnosis of blast furnaces. The system analysis more than a hundred parameters in real time and shows the operator the status of the blast furnace and its behaviour. The system also gives suggestions and control set-points in order to achieve a better hot metal quality and aimed analysis, as well as to increase the production in the blast furnace. In addition to the information given by the furnace, the expert system also takes advantage of a theoretical model of behaviour of thermal status and permeability and apply the considerations given by the human experts in order to solve all the conflicts.

Test Construction and Management with Network Adaptive Control

Testing by computer is becoming more and more important in the present Tutoring Systems and Distance Learning Models. In this work, starting from several previous contributions and considering the assessment methodologies specified mainly by the Institute of Education Sciences of the University of Oviedo, an integrated model of test processing is proposed consisting of the following elements: an open items editor compatible with the current word processors that permits the highest flexibility at the moment of composing the enunciations and distracters with different multimedia elements, an items classification system in thematic areas making sorting and content balancing feasible in test working and facilitating the portability in different environments, a mechanism of selection and presentation to the student that permits simultaneously teacher predefined tests, a random access to the items for training and a self-assessment bayesian system, and finally, by means of a Client/Server architecture, an application for net management that facilitates their use as much in local area networks (LANs) as extended through the Internet.

Approach to Intelligent Adaptive Testing

Considering the evaluation of acquired knowledge as a fundamental element of Computerized Tutoring Systems, the classical Bayesian methods for adaptive testing present several deficiencies, such as arbitrary assignments or prior probabilities, non representativity, the effect of the context and the need to include all the concepts under evaluation. An alternative model of the adaptive test is proposed here based on content-balanced and imprecise interpretations of the information level and prior distribution of each item using Gil”s fuzzy relationships instead of Bayesian relationships.

Foundations of an Adaptative Tutoring System Based on Systemic Networks

In this paper, the elements which take part in a Tutoring System: knowledge organization, concepts, versions, evaluation, student typology and their activity, are considered as constituting a determined language: the teaching language.