Abstract
In data imputation problems, researchers typically use several techniques, individually or in combination, in order to find the one that presents the best performance over all the features comprised in the dataset. This strategy, however, neglects the nature of data (data distribution) and makes impractical the generalisation of the findings, since for new datasets, a huge number of new, time consuming experiments need to be performed. To overcome this issue, this work aims to understand the relationship between data distribution and the performance of standard imputation techniques, providing a heuristic on the choice of proper imputation methods and avoiding the needs to test a large set of methods. To this end, several datasets were selected considering different sample sizes, number of features, distributions and contexts and missing values were inserted at different percentages and scenarios. Then, different imputation methods were evaluated in terms of predictive and distributional accuracy. Our findings show that there is a relationship between features’ distribution and algorithms’ performance, and that their performance seems to be affected by the combination of missing rate and scenario at state and also other less obvious factors such as sample size, goodness-of-fit of features and the ratio between the number of features and the different distributions comprised in the dataset. © Springer Nature Switzerland AG 2018.

Abstract

Abstract

Abstract

Abstract
This article serves as an introduction to the Special Issue on Metalearning and Algorithm Selection. The introduction is divided into two parts. In the the first section, we give an overview of how the field of metalearning has evolved in the last 1-2 decades and mention how some of the papers in this special issue fit in. In the second section, we discuss the contents of this special issue. We divide the papers into thematic subgroups, provide information about each subgroup, as well as about the individual papers. Our main aim is to highlight how the papers selected for this special issue contribute to the field of metalearning.

Abstract
This paper describes the collaborative learning environment, aligned with the United Nations Millennium Development Goals, provided by the European Project Semester (EPS). EPS is a one semester capstone project programme offered by eighteen European engineering schools as part of their student exchange programme portfolio. In this international programme, students are organized in teams, grouping individuals from diverse academic backgrounds and nationalities. The teams, after choosing a project proposal, become fully responsible for the conduction of their projects. By default, project proposals refer to open multidisciplinary real problems. The purpose of the project is to expose students to problems of a greater dimension and complexity than those faced throughout the degree programme as well as to put them in contact with the so-called real world, in opposition to the academic world. EPS provides an integrated framework for undertaking capstone projects, which is focused on multicultural and multidisciplinary teamwork, communication, problem-solving, creativity, leadership, entrepreneurship, ethical reasoning and global contextual analysis. Specifically, the design and development of sustainable systems for growing food allow students not only to reach the described objectives, but to foster sustainable development practices. As a result, we recommend the adoption of this category of projects within EPS for the benefit of engineering students and of the society as a whole.