Javier López-Cuadrado

Calibration of an item bank for the assessment of Basque language knowledge

The main requisite for a functional computerized adaptive testing system is the need of a calibrated item bank. This text presents the tasks carried out during the calibration of an item bank for assessing knowledge of Basque language. It has been done in terms of the 3-parameter logistic model provided by the item response theory. Besides, this work focuses on the most important decisions taken and the algorithms followed during the process. Actually, the most significant novelties introduced to the traditional calibration procedure deal with the adaptation of the algorithm and the putting into practice of both supervised and non-supervised web-based subtest administrations. Once the item bank had been calibrated, it was added to the generator of computerized adaptive admission tests that is expected to be operative in the new version of Hezinet, a multimedia Basque language e-learning management system that is being used since 2000 in more than 60 Basque-adult schools.

On Evaluating Adaptive Systems for Education

In this paper we have gathered some interesting techniques to evaluate educative systems. Our main purpose is to evaluate HEZINET, an adaptive hypermedia system for education available commercially. Therefore, we also include a system evaluation plan to identify which types and techniques of evaluation will be needed in order to accomplish it.

Automatic construction of Fuzzy Inference Systems for computerized clinical guidelines and protocols

We create Fuzzy Inference Systems (FIS) as a means of computerizing differential diagnosis (DD) tables.Use of Model-Driven Software Engineering (MDSE) techniques to systematize FIS development process.FIS design and edition from both domain expert and knowledge engineer perspectives.Calculation results shown in an easily comprehensible and self-explanatory way.Tested in the development of two FIS for a computerized clinical guideline for hyperammonemia care. Clinical guidelines and protocols (CGPs) are standard documents with the aim of helping practitioners in their daily work. Their computerization has received much attention in recent years, but it still presents some problems, mainly due to the low sustainability and low adaptability to changes (both in knowledge and technology) of the computerized CGPs. This paper presents an approach to an easy and automatic creation of Fuzzy Inference Systems (FISs), which are suitable for the computerized interpretation of differential diagnoses. The proposed FIS development process is based on applying Model-Driven Software Engineering techniques: automatic generation of computer artefacts and separation of concerns. The process focuses on the separation of roles during the design stage: domain experts use a basic editor that allows them to define the categories and factors that will be involved in the FIS in natural language, while knowledge engineers at a later stage refine these elements using a more advanced editor. The whole system has been tested by automatically generating two FISs that have been included in a computerized CGP for the diagnosis of a rare disease called hyperammonemia. This CGP has been validated and it is currently in use.

Some Innovations for Keeping an E-Learning System Alive

Keeping an e-learning system alive consists not only in developing and installing it, but also in maintaining the system updated, improving its features and adding new ones. Hezinet is a commercial e-learning system for Basque language learning that is being used since 2000 in more than 60 Basque-adult schools. In this period, although the good results achieved, authors have realized the necessity of some improvements and new features. This paper introduces three new components of Hezinet as a way to present some key issues in the development of a successful e-learning tool: a computer adaptive admission test generator, an on-line calibration supporting component, and an item/test authoring tool.