Abstract
Governmental investments on the development of high-tech clusters are among the main policies for socioeconomic development, enabling countries to be part of global value networks. Our objective is to identify which are the strategies of countries that want to join global aerospace value networks, by means of an abductive case research. Countries were divided in 3 groups (A; B: C) according to their global aerospace exports share. The analytical framework used to identify the strategies has 3 dimensions: network structure, network governance, and network dynamics. Results show different strategies according to the country's global exports share. While for countries in group A (exports above 1%), a strategy focused on the dimension network structure indicated a sustained high-tech sector. Countries in group C tend to focus on specialization, taking advantage of shifts in technological paradigms to upgrade their development level. The dimension network governance is mainly related to governmental efforts toward the creation of clusters and associations, promoting specialization and collaborative work. Finally, the dimension network dynamics describes the attraction of foreign companies to qualify the clusters at countries who belong to group C, while countries at group A reinforce their research and development activities. The comparison between countries is helpful for governmental representatives who want to develop strategies toward increasing participation in an industrial global value network and for supply chain managers to help selecting the locations for their operations.

Abstract
Purpose Based on the data obtained from a questionnaire of 595 people, the authors explore the relative importance of consumers, checking whether socioeconomic variables influence their centrality, detecting the communities within the network to which they belong, identifying consumption patterns and checking whether there is any relationship between co-marketing and consumer choices. Design/methodology/approach A multilayer network is created from data collected through a consumer survey to identify customers' choices in seven different markets. The authors focus the analysis on a smaller kinship and cohabitation network and apply the LART network community detection algorithm. To verify the association between consumers' centrality and variables related to their respective socioeconomic profile, the authors develop an econometric model to measure their impact on consumer's degree centrality. Findings Based on 595 responses analysing individual consumers, the authors find out which consumers invest and which variables influence consumers' centrality. Using a smaller sample of 70 consumers for whom they know kinship and cohabitation relationships, the authors detect communities with the same consumption patterns and verify that this may be an adequate way to establish co-marketing strategies. Originality/value Network analysis has become a widely used technique in the extraction of knowledge on consumers. This paper's main (and novel) contribution lies in providing a greater understanding on how multilayer networks represent hidden databases with potential knowledge to be considered in business decisions. Centrality and community detection are crucial measures in network science which enable customers with the highest potential value to be identified in a network. Customers are increasingly seen as multidimensional, considering their preferences in various markets.

Abstract

Abstract
In recent years, the development and use of mobile devices such as smartphones and tablets grew significantly. They are used for virtually every activity of our lives, from communication or online shopping to e-banking or gaming, just to name a few. As a consequence, these devices contribute significantly to make our lives more digital, with all the perks and risks that this encompasses. One of the most serious risk is that of an authorized individual gaining physical access to our mobile device and, potentially, to all the applications and personal data it contains. Most of mobile devices are protected using some kind of password, that can be easily spotted by unauthorized users or event guessed. In the last years, new authentication mechanisms have been proposed, such as those using traditional biometrics or behavioral biometrics. In this paper we propose a new continuous authentication mechanism for mobile devices based on behavioral biometrics that monitors user interaction behavior for classifying the identity of the user.

Abstract
Entropy models the added information associated to data uncertainty, proving that stochasticity is not purely random. This paper explores the potential improvement of machine learning methodologies through the incorporation of entropy analysis in the learning process. A multi-layer perceptron is applied to identify patterns in previous forecasting errors achieved by a machine learning methodology. The proposed learning approach is adaptive to the training data through a re-training process that includes only the most recent and relevant data, thus excluding misleading information from the training process. The learnt error patterns are then combined with the original forecasting results in order to improve forecasting accuracy, using the Rényi entropy to determine the amount in which the original forecasted value should be adapted considering the learnt error patterns. The proposed approach is combined with eleven different machine learning methodologies, and applied to the forecasting of electricity market prices using real data from the Iberian electricity market operator – OMIE. Results show that through the identification of patterns in the forecasting error, the proposed methodology is able to improve the learning algorithms’ forecasting accuracy and reduce the variability of their forecasting errors.

Abstract
The rapid increase of connected devices causes more and more data to be generated and, in some cases, this data needs to be analyzed as it is received. As such, the challenge of presenting streaming data in such way that changes in the regular flow can be detected needs to be tackled, so that timely and informed decisions can be made. This requires users to be able to perceive the information being received in the moment in detail, while maintaining the context. In this paper, we propose VisMillion, a visualization technique for large amounts of streaming data, following the concept of graceful degradation. It is comprised of several different modules positioned side by side, corresponding to different contiguous time spans, from the last few seconds to a historical view of all data received in the stream so far. Data flows through each one from right to left and, the more recent the data, the more detailed it is presented. To this end, each module uses a different technique to aggregate and process information, with special care to ensure visual continuity between modules to facilitate the analysis. VisMillion was validated through a usability evaluation with 21 participants, as well as performance tests. Results show that it fulfills its objective, successfully aiding users to detect changes, patterns and anomalies in the information being received.