Abstract
Tourism crowdsourcing platforms accumulate and use large volumes of feedback data on tourism-related services to provide personalized recommendations with high impact on future tourist behavior. Typically, these recommendation engines build individual tourist profiles and suggest hotels, restaurants, attractions or routes based on the shared ratings, reviews, photos, videos or likes. Due to the dynamic nature of this scenario, where the crowd produces a continuous stream of events, we have been exploring stream-based recommendation methods, using stochastic gradient descent (SGD), to incrementally update the prediction models and post-filters to reduce the search space and improve the recommendation accuracy. In this context, we offer an update and comment on our previous article (Veloso et al., 2019a) by providing a recent literature review and identifying the challenges laying ahead concerning the online recommendation of tourism resources supported by crowdsourced data.

Abstract
The emergence of online social networks provided users with an easy way to publish and disseminate content, reaching broader audiences than previous platforms (such as blogs or personal websites) allowed. However, malicious users started to take advantage of these features to disseminate unreliable content through the network like false information, extremely biased opinions, or hate speech. Consequently, it becomes crucial to try to detect these users at an early stage to avoid the propagation of unreliable content in social networks’ ecosystems. In this work, we introduce a methodology to extract large corpus of unreliable posts using Twitter and two databases of unreliable websites (OpenSources and Media Bias Fact Check). In addition, we present an analysis of the content and users that publish and share several types of unreliable content. Finally, we develop supervised models to classify a twitter account according to its reliability. The experiments conducted using two different data sets show performance above 94% using Decision Trees as the learning algorithm. These experiments, although with some limitations, provide some encouraging results for future research on detecting unreliable accounts on social networks. © 2020, Springer Nature Switzerland AG.

Abstract
Evermore end-to-end digital services depend on the proper interoperation of multiple products, forming a distributed system, often subject to timing requirements. To ensure interoperability and the timely behavior of such systems, it is important to conduct integration tests that verify the interactions with the environment and between the system components in key scenarios. The automation of such integration tests requires that test components are also distributed, with local testers deployed close to the system components, coordinated by a central tester. Test coordination in such a test architecture is a big challenge. To address it, in this article we propose an approach based on the pre-processing of the test scenarios. We first analyze the test scenarios in order to check if conformance errors can be detected locally (local observability) and test inputs can be decided locally (local controllability) by the local testers for the test scenario under consideration, without the need for exchanging coordination messages between the test components during test execution. If such properties do not hold, we next try to determine a minimum set of coordination messages or time constraints to be attached to the given test scenario to enforce those properties and effectively solve the test coordination problem with minimal overhead. The analysis and enforcement procedures were implemented in the DCO Analyzer tool for test scenarios described by means of UML sequence diagrams. Since many local observability and controllability problems may be caused by design flaws or incomplete specifications, and multiple ways may exist to enforce local observability and controllability, the tool was designed as a static analysis assistant to be used before test execution. DCO Analyzer was able to correctly identify local observability and controllability problems in real-world scenarios and help the users fix the detected problems.

Abstract
In recent years, the large deployment of distributed energy resources (DERs) in low voltage networks is changing the traditional approach to power systems. This massive change is pushing towards new solutions to improve energy trading in low voltage networks. Consumer-centric options, such as full peer-to-peer (P2P) and energy community markets (CM) are seen as viable options to increase the active participation of end-users in the electricity markets. This work studies the full P2P and CM market approaches applied to the actual regulatory framework in Brazil, evaluating and comparing both approaches to be potentially applied in Brazil. A case study based on a typical Brazilian neighborhood is designed, allowing to assess the behavior of consumers and prosumers in both markets. The results show the economic viability of both models, considering the social welfare and the penetration of distributed generation in the system. An important conclusion under the current regulatory framework is that the full P2P can have greater benefits over the CM, as long as the distributed generation is enough to confer near self-sufficiency to the peer's demand.

Abstract
Application execution required in cloud and fog architectures are generally heterogeneous in terms of device and application contexts. Scaling these requirements on these architectures is an optimization problem with multiple restrictions. Despite countless efforts, task scheduling in these architectures continue to present some enticing challenges that lead us to question how tasks are routed between different physical devices, fog nodes and cloud. In fog, due to its density and heterogeneity of devices, the scheduling is very complex and, in the literature, there are still few studies that have been conducted. However, scheduling in the cloud has been widely studied. Nonetheless, many surveys address this issue from the perspective of service providers or optimize application quality of service (QoS) levels. Also, they ignore contextual information at the level of the device and end users and their user experiences. In this paper, we conducted a review of the literature on the main task scheduling algorithms in cloud and fog architecture; we studied and discussed their limitations, and we also explored and suggested some perspectives for improvement.

Abstract
The use of graphene oxide-based coatings on optical fibers are investigated, aiming to tune the reflectivity of optical fiber surfaces for use in precision sensing devices. Graphene oxide (GO) layers are successfully deposited onto optical fiber ends, either in cleaved or hollow microspheres, by mounting combined bilayers of polyethylenimine (PEI) and GO layers using the Layer-by-Layer (LbL) technique. The reflectivity of optical fibers coated with graphene oxide layers is investigated for the telecom region allowing to both monitor layer growth kinetics and cavity characterization. Tunable reflective surfaces are successfully attained in both cleaved optical fibers and hollow microsphere fiber-based sensors by simply coating them with PEI/GO layers through the LbL film technique.