Abstract
With the profusion of web content, researchers have avidly studied and proposed new approaches to enable the anticipation of its impact on social media, presenting many distinct approaches throughout the last decade. Diverse approaches have been presented to tackle the problem of web content popularity prediction, including standard classification and regression approaches. Furthermore, these approaches have also taken into consideration distinct scenarios of data availability, where one may target the prediction of popularity before or after the publication of the items, which is highly interesting for different objectives from a user standpoint. This work aims at reviewing previous work and discussing open issues and challenges that could foster impactful research on this topic. Five areas are identified that require further research, covering the full spectrum of the problem: social media data, the learning task, recommendation and evaluation. © 2019 Elsevier B.V.

Abstract
In this paper the proposal of a low cost high performance educational mobile robot is described. The robot is based on an Arduino, applied in the low level control, while the high level control loop is carried out by an RPI running an object pascal application. The described robot was prototyped in order to have a competitive participation in the Robotic Day Line Follower 2017 competition, taking advantage of the RPI capabilities. The RPI allows the use of higher performance sensors, when compared with the most common standard approaches based on a single 8 bit RISC micro-controller, having as disadvantage the inevitable robot size increase, which compromises in certain situations the robot maneuverability and increases the power consumption. The robot is equipped with DC Motors, the chosen line follower sensor is the picamera and for the obstacle detection sonar sensors are used. © CLAWAR Association.

Abstract
Searching for health information is the third most popular activity on the Internet. There is evidence that query suggestions in lay and medico-scientific terminology improve health information retrieval by who is not a health professional. Developing systems that suggest queries in these terminologies requires knowing if concepts are lay or medico-scientific. In this paper, we propose and compare approaches to compute the degree of association of a concept to lay and medico-scientific terminology. We use different thesauri for this purpose and use the cosine similarity to measure the closeness of concepts with subsets of those thesauri. The evaluation of our approaches uses an existing glossary containing concepts in both terminologies in English and Portuguese and a and a set of queries submitted by users and classified by health professionals as lay or medical-scientific. We concluded that the best method to classify a concept uses the CHV vocabulary as a subset.

Abstract
Since the emergence of the Global Village, the information flow changed drastically. Digital Technologies changed how people communicate, how they access information and how they share it. It gave people an unlimited exposure to information and knowledge. However, it also seemed to limit it. Recommendation algorithms are used in order to provide a customized experience that captivates users. Although they play an important role in selecting information that is considered relevant to the user, significant information/content may be omitted. Consequently, users end up closed in a bubble of limited information, which affects critical thinking skills and appears to influence and guide personal opinions. Little attention has been given to the negative effects of information bias on people’s critical thinking. Thus, it is hoped that this study will at the same time educate and bring awareness to this issue. In a survey we performed (with 117 answers) the majority of the survey sample (approximately 54,7%) revealed discomfort regarding the storage and filtering of data. Interestingly, 29,9% of the participants were found to be indifferent regarding this issue. From these results, the authors can conclude that, although most of the participants feel uncomfortable, they prefer to be passive about this, which reinforces the idea of conformity and the false sense of organization mentioned herein. An interview with an expert in the area drew attention to the fact that social pressure most often leads users to comply and rely on the group’s beliefs and attitudes, which facilitates social relationships and avoids confrontation.

Abstract
Population in the cities grows every day posing new challenges that need to be properly addressed throughout the planning and design of sustainable and smart cities - cities of the future. Mobility is an important issue considering social inclusion and the sustainable development of such cities. Thus, future mobility will have an increased importance when having to plan and design the cities of tomorrow. A key component of any future mobility and its metabolism is what is known as Mobility as a Service (MaaS), representing emerging opportunities from any type or mode of transportation in future cities. Through an empirical and explorative research methodology, this paper presents the main APP's/platforms characteristics, regarding the European territory, towards an integrated and sustainable mobility. Concluding, we present some features and trends that should be considered in the development of future MaaS systems, allowing a more convenient provision of versatile and attractive mobility services. © 2018 IEEE.

Abstract
There is no standard definition of outliers, but most authors agree that outliers are points far from other data points. Several outlier detection techniques have been developed mainly for two different purposes. On one hand, outliers are considered error measurement observations that should be removed from the analysis, e.g. robust statistics. On the other hand, outliers are the interesting observations, like in fraud detection, and should be modelled by some learning method. In this work, we start from the observation that outliers are affected by the so-called simpson paradox: a trend that appears in different groups of data but disappears or reverses when these groups are combined. Given a data set, we learn a regression tree. The tree grows by partitioning the data into groups more and more homogeneous of the target variable. At each partition defined by the tree, we apply a box plot on the target variable to detect outliers. We would expect that the deeper nodes of the tree would contain less and less outliers. We observe that some points previously signalled as outliers are no more signalled as such, but new outliers appear.