Ainhoa Álvarez

The IRIS Authoring Tool

IRIS is an authoring tool developed to help human instructors to build intelligent teaching-learning systems in a wide variety of domains. The instructor/designer is required to produce a set of pedagogical requisites, which, in turn, are used to automatically configure a generic tutor architecture and produce an intelligent tutoring system. In order to provide IRIS with a sound basis for producing systems, a theory of instruction that integrates cognitive processes, instructional events and instructional actions has been considered. In this chapter we explain, through the analysis of the cognitive theory, the generic architecture. We present several system components, the requirements of the different components, and show how we integrate these components in IRIS. Moreover, we embed requirements, cognitive principles, and design requisites in an authoring tool in order that human instructors can follow them easily. Various design issues, an example of building a tutor for mathematical differentiation by using IRIS, and some experiences of use and evaluation are also presented.

Helping Teachers to Track Student Evolution in a B-Learning Environment

SigMa is an adaptable feedback generation tool for the three kinds of actors involved in a blended-learning process: teachers, students, and the e-learning platform. This paper focuses on the teacher tool and defines its desirable functionalities and main components. The teacher tool gathers data provided by the e-learning platform’s diagnostic capabilities and analyses it. The results are shown to teachers by means of a Visualisation Interface that informs them about course development and suggests possible changes and improvements. A Teacher Profile is used to adapt the feedback to match teachers’ strategies and preferences.

Experiences Incorporating Lego Mindstorms Robots in the Basic Programming Syllabus: Lessons Learned

Basic Programming is a first year mandatory course of the Computer Engineering degree. Both students and teachers face difficulties in this course, which has high failure and drop-out rates. Several authors have proposed the use of visual programming environments and robots to overcome the difficulties of this course, some of which have been successful. This paper presents the two-year experiment using Lego Robots carried out at the University of the Basque Country (UPV/EHU) with around 100 students, along with the results. Satisfactory results have been obtained regarding both motivation and the perception of the students of their learning process; moreover the drop-out rate decreased even though no statistical significance was obtained regarding the final marks of the course. From those results and the analysis of the data it was derived that robot sessions should be more integrated in the curriculum, giving them greater relevance in the final marks. In addition, it is indispensable to classify course students and adapt learning sessions to each student type due to the high student heterogeneity.

Dealing with common problems in engineering degrees' Final Year Projects

After the implementation of the Bologna Process, the Final Year Project (FYP) is a mandatory course in all bachelor degrees in Spanish higher education system. Although the FYP is new for many degrees, it was already established in the previous education system for engineering degrees. The authors of this paper are lecturers in a bachelor degree in Computer Science at three different colleges of the University of the Basque Country UPV/EHU. In the last few years, they have supervised more than 80 FYPs and they have all faced similar problems during the supervision process. This paper presents a study conducted to determine if the observed problems are just a perception of the authors or whether they are general problems in the elaboration of FYPs. In the study, 60 former students and 27 lecturers took part. Four main categories of problems have been identified and a proposal to overcome them is presented.

Generating Teacher Adapted Suggestions for Improving Distance Educational Systems with SIgMa

In this paper we present SIgMa, an adaptable feedback generation tool for teachers which gives information about course development in an educational Web environment with diagnostic capabilities. First student results are analysed using statistical calculations and data mining techniques. Afterwards, the feedback provision is carried out by a rule-based system that recognizes anomalous behaviour patterns among the analysis results and provides some suggestions for improvement which are adapted to teacher strategies and preferences by means of a teacher model. We present the main proposal and the implementation and a test of a restricted prototype.

Incorporating educational robots and visual programming environments in introductory programming courses

Introductory programming courses are very challenging both for students and teachers. Several authors propose incorporating either educational robots or visual programming environments as a means to ease learning of programming. This paper presents three experiences that use these kinds of tools with the aim of applying Kolbs experiential learning cycle. The experiences have been carried out in the first year courses of programming in the Bachelor in Computer Management and Information Systems Engineering at the Faculty of Engineering of Vitoria-Gasteiz in the UPV/EHU.

Experiences in Visualizing the Analysis of Blended-Learning Interactions to Support Teachers

In the context of the MAgAdI environment, this paper presents SIgMa, a teacher oriented tool, and SIgBLE, its general framework basis. SIgMas general aims are to automatically detect visible signs of failure or success among data coming from interactions happened on blended learning environments, and to provide relevant feedback adapted to teachers. First, we introduce SIgBLE and the interface of the SIgMa prototype. A usability test of the current prototype is also presented.

Focusing on Personal Organization to Enhance Overall E-learning

This paper describes the inclusion of a recommendation system in the MAGADI [1] blended-learning environment. It seeks to help students with their overall study process by recommending both which courses to study for and the contents to be developed on each

Adaptive learning based on variable student and domain models in Magadi

Magadi is an online multidomain agent based adaptive educational environment developed on the conceptual results obtained in the IRIS system. In this paper, we propose the extension of Magadi by including a teacher model and the possibility of using different student and domain views. Besides, a mechanism to adaptively select the planning rules to be used in the generation of the student adapted learning paths is provided.

Supporting complex assessment processes

Assessment is a key element in the teaching-learning process. Every formal education implies some kind of assessment, and the results of this assessment contribute to the quality improvement in that process. Currently, most assessment scenarios are very complex and require tools to support them. Establishing a good assessment approach requires continuous decision-making processes to guarantee fair assessments. In this paper the TraMeA assessment support system is presented centering on its characteristics that can help in those decision making processes. The proposal is exemplified with real examples derived from the evaluation of Final Year Projects.