Software Engineering and Artificial Intelligence

In this paper, we propose three different ways of using IMS Learning Design to support online adaptive learning modules that include educational videogames. The first approach relies on IMS LD to support adaptation procedures where the educational games are considered as Learning Objects. These games can be included instead of traditional content to adapt the learning experience to the preferences of the student. In the second approach, the game engine supports the entire adaptation cycle, simply using IMS LD as a delivery mechanism for the packages containing the games. The learner profile is used to adapt the game, and the activity within the game is used to adjust the profile. Finally, the third approach stands between the other two, using IMS LD to model the general adaptation guidelines and then refining the adaptation inside the games. We illustrate these approaches with the <e-Adventure> educational game engine, which includes the mechanisms required to support these adaptation cycles.

Creating web based educational (e-learning) applications is a complex process that poses several challenges. We are especially concerned with two aspects: the cost of production and maintenance of educational contents, and the personalization of contents to suit individual learners. The Learning Objects (LO) model was proposed in response to reusability and adaptability concerns. IMS, ADL, IEEE and other initiatives have proposed standards for metadata description and packaging facilities in order to facilitate the interchange of LOs across Learning Management Systems (LMS). Metadata and packaging description demand additional work from content developers. In this paper, we address how DocBook, an XML/SGML document model, facilitates the creation of learning content and allows the countless DocBook materials already existing on the Internet to be reused. The main goals are to simplify the creation of adaptive learning content following the LO paradigm and to study how DocBook and its associated tools facilitate the authors’ task when dealing with the XML documents needed for LO description and interchange.

In this paper we present a general framework, called NUCLEO, for the application of socio-constructive educational approaches in higher education. The underlying pedagogical approach relies on an adaptation model in order to improve group dynamics, as this has been identified as one of the key features in the success of collaborative learning scenarios. Students' learning strategies are analyzed considering a simplified version of Vermunt's model for learning styles. The resulting profiles provide the basis to group students in teams where each student is assigned a role according to his/her learning strategies. The result is the formation of complementary and semi-autonomous learning teams that collaborate to achieve solutions to the problems provided by the instructor. The framework is instantiated through an online multiplayer role-playing game environment, which sets a stage for the underlying collaborative problem-based learning approach. The framework has been benchmarked in different programming courses at the Complutense University of Madrid during 2007-08 and 2008-09 academic years. This work also presents the results of these experiences.

University courses about computer programming usually seek to provide students not only with technical knowledge, but also with the skills required to work in real-life software projects. Nowadays, the development of software applications requires the coordinated efforts of the members of one or more teams. Therefore, it is important for software professionals to master the sort of skills that assure the success of teamwork, such as communication, leadership, negotiation, or team management. However, these abilities are difficult to teach, one of the reasons being that they require true commitment from the students. However, today students are taking a more and more passive role in their own education, two of the more evident consequences being the increase in dropout rates and the decrease in marks obtained in exams. The NUCLEO e-learning framework has been designed to promote the effective acquisition of teamwork skills and, at the same time, to promote the more active participation of the students in their own learning process. NUCLEO adopts a socio-constructivist pedagogical approach that pursues the development of communities of practice for Problem Based Learning. Our research has rooted the design decisions of NUCLEO in the analysis of its socio-cultural environment with Activity Theory, which considers conflicts within groups as the impetus of their evolution and the forges of their environments. This paper presents the analysis of the main features of NUCLEO according to Activity Theory, as well as the experimental results obtained with the framework in three different case studies in university courses.

If citing, it is advised that you check and use the publisher's definitive version for pagination, volume/issue, and date of publication details. And where the final published version is provided on the Research Portal, if citing you are again advised to check the publisher's website for any subsequent corrections.

Software projects estimations are a crucial component of successful software development. There have been many approaches that deal with this problem by using different kinds of techniques. Most of the successful techniques rely on one shot prediction of some variables, as cost, quality or risk, taking into account some metrics. However, these techniques usually are not able to deal with uncertainty on the data, the relationships among metrics or the temporal aspect of projects. During the last decade, some researchers have proposed the use of Bayesian Belief Networks (BBNs) to perform better estimations, by explicitly taking into account the previous shortcomings. But, these approaches were based on manually defining those BBNs and handling only one of the estimation variables (cost, quality or risk). In this paper, we present an approach for semi-automatically building BBNs by using machine learning techniques. We describe two algorithms to generate such BBNs. The first one generates one-shot BBNs, while the second one generates BBNs that take into account the temporal aspect of project development. We performed experiments on real data coming from two software companies, obtaining a 63% of accuracy on multiclass classification. Our main interest was to find a semantically correct model that can be trained with future projects to increase its accuracy. In this sense, we introduce a well-balanced approach to make good predictions with strong explanatory power.

The paper describes recent work for experimenting on-board autonomy capabilities within a larger multi-agent system that provides product delivery for the GMES Earth Observation System. We will show how robust state of the art technology for constraint-based scheduling and execution can be customized for endowing agents with the autonomy capability.

If citing, it is advised that you check and use the publisher's definitive version for pagination, volume/issue, and date of publication details. And where the final published version is provided on the Research Portal, if citing you are again advised to check the publisher's website for any subsequent corrections.

The project CoRob-X develops and demonstrates enabling technologies for multi-agent robotic teams to explore planetary surfaces with a focus on hard-to-reach areas where a collaborative scheme is required to efficiently explore complex environments. Exploring lava tubes is such a challenging environment and requires a team of robots able to collaborate in an autonomous way to find their way to the subsurface tube system, descend through a natural entry hole (the so-called skylight), and explore the interior with payload instruments to provide scientific data. The developed robotic exploration system that will tackle the ambitious goal is composed of three rovers with substantially different technical characteristics. The paper presents the overall approach, i.e., the control architecture, the robotic systems, and the software to be used. It also showcases the selected mission phases that will be demonstrated in a field-test campaign. In addition, a terrestrial mining use case is presented that demonstrates how the developed autonomy-enabling software can be transferred to terrestrial applications.

This document constitutes the MARTES project deliverable D1.1 "Current limitations of best practices". The objective is to analyze the limitations of selected relevant methodology approaches in order to focus the MARTES research work on profound methodological issues, and to learn from the experiences and solutions experimented in the best practices. Each of the contributors to the deliverable has analyzed one or more approaches either based on their knowledge of using the methodology or on careful studies of published papers and documents. Chapter 1 summarizes the expected features and properties of the MARTES methodology. Chapter 2 describes the analysis results of 10 relevant approaches. Chapter 3 describes approaches and needs of participating companies. Chapter 4 deals with approaches that university and research partners have used. Chapter 5 presents a summary of observations. ITEA 04006 Current limitations of best practices Deliverable ID: 1.1