Abstract
Imbalanced domains are an important problem frequently arising in real world predictive analytics. A significant body of research has addressed imbalanced distributions in classification tasks, where the target variable is nominal. In the context of regression tasks, where the target variable is continuous, imbalanced distributions of the target variable also raise several challenges to learning algorithms. Imbalanced domains are characterized by: (1) a higher relevance being assigned to the performance on a subset of the target variable values; and (2) these most relevant values being underrepresented on the available data set. Recently, some proposals were made to address the problem of imbalanced distributions in regression. Still, this remains a scarcely explored issue with few existing solutions. This paper describes three new approaches for tackling the problem of imbalanced distributions in regression tasks. We propose the adaptation to regression tasks of random over-sampling and introduction of Gaussian Noise, and we present a new method called WEighted Relevance-based Combination Strategy (WERCS). An extensive set of experiments provides empirical evidence of the advantage of using the proposed strategies and, in particular, the WERCS method. We analyze the impact of different data characteristics in the performance of the methods. A data repository with 15 imbalanced regression data sets is also provided to the research community.

Abstract
The search for logistics best-practices in international trade has led to the appearance of the Berth Allocation Problem. If the vessels have release dates, the problem is proved to be NP-hard and the performance of exact algorithms is not satisfactory, leading to the use of metaheuristics. This paper develops a Hybrid Evolutionary Algorithm for the discrete and dynamic Berth Allocation Problem. A challenge of using Genetic Algorithms is the identification of the best approach to model a specific problem. This paper proposes the use of frontier techniques (Data Envelopment Analysis and Free Disposal Hull models) to compare the performances of alternative specifications of the parameters for the algorithm proposed and to identify efficient solutions.

Abstract
Smartphones are becoming more important in our everyday lives and it is increasingly common to perform critical tasks on these devices, such as making payments. For this reason, ensuring the quality of these applications is an important task. One way to do this is through software testing. However, the testing of these applications presents major challenges due to the wide variety of devices available in the market. In this context, automated testing gains more relevance. There are dynamic test approaches for testing mobile applications, but there are some challenges that need to be overcome for good results, such as, being able to explore the complete behaviour of the application (e.g., overcoming blocking points); choosing appropriate input data; testing dynamic behaviour; testing specific characteristics of mobile applications, such as specific forms of interaction, e.g., long press, and so on. This paper presents a dynamic exploration approach of Android mobile applications that aims to overcome some of the problems identified. During the exploration process, the algorithm builds a Finite State Machine where states are traversed screens and transitions between states describe events that allow moving from one screen to another. This approach is implemented as an extension of the iMPAcT tool. The approach is validated over real Google Play apps and the test coverage results achieved are presented, compared and discussed.

Abstract
The development of efficient methods for searching and browsing large assets of video content has been considered by the academia and content owners for long. Different approaches that range from manual structured annotations, to unstructured metadata collected from several sources, as well as multimedia processing for automatic description of the content, can be identified. The growth on the number of hours of video content put online in video sharing platforms has however shown that video retrieval is still quite inefficient as rich contextual data that describes the content is most of the times still not available. Additionally, metadata is usually not linked to timed moments of content, making direct access to the most relevant moments not possible. In this paper, an approach for making web videos available in the YouTube platform more accessible is presented. The solution is based on a collaborative process presented as a game that enables collecting metadata from the crowd while implementing mechanisms that remove erroneous information usually encountered in this type of information. Metadata, exported to YouTube in the form of captions and descriptions, contributes to enhance video retrieval, guaranteeing a better user experience and exposure of the content.

Abstract

Abstract
Over the last decade, for those teaching marketing, it has been a decade of profound change. The importance of digital marketing changed a lot for organizations in this period. WhatsApp had just been founded, in February 2009; but was very far from its over 1 billion users, in more than 180 countries, of today. Instagram did not exist yet. The iPhone was less than two years old. Smartphones were still not accessible to everyone, nor were there so many apps to help solve our day-to-day problems. This article is an application of autoethnography and explains the emotional journey as a marketing lecturer that the author has experienced between 2009 and 2019. It is thus a personal account, based on the experience of the author, about the change which has occurred in the classroom, as well as outside it, during that period. Higher education is an intangible service which has been very affected by technology, the result being a culture with greater proximity and where students seek to have positive experiences.