Abstract
Service applications are increasingly being deployed in virtualized environments, such as virtual machines (VMs) as a means to provide elasticity and to allow fast recovery from failures. The recent trend is now to deploy applications in containers (e.g., Docker or RKT containers), which allow, among many other benefits, to further reduce recovery time, since containers are much more lightweight than VMs. Although several performance benchmarks exist for web services (e.g., TPC-App and SPEC SPECjEnterprise2010) or even virtualized environments (e.g., SPEC Cloud IaaS 2016, TPCx-V), understanding the behavior of containerized services in the presence of faults has been generally disregarded. This paper proposes an experimental approach for evaluating the performance of containerized services in presence of operator faults. The approach is based on the injection of a simple set of operator faults targeting the containers and middleware. Results show noticeable differences regarding the impact of operator faults in Docker and RKT, with the latter one allowing for faster recovery, despite showing the lowest throughput.

Abstract
In a digital forensic investigations, the investigator usually wants to get as much state information as possible. Examples of such scenarios are households with wireless networks connecting multiple devices where a security incident occurs. USB devices present themselves as interesting vehicles for the automated collection of state information, as it can store the applications that collect the information, can store the results and can also facilitate the information collection by enabling its automatic operation. This paper proposes a USB solution to facilitate the collection of state information with integrity guarantees and multi-platform operation. Moreover, the proposed solutions is the only one that performs an extensive and homogeneous artifact collection, independently of the underlying operating system.

Abstract
The information generated by a network monitoring system is overwhelming. Monitoring is imperative but very difficult to accomplish due to several reasons. More so for the case of non tech-savvy home users. Security Information Event Management applications generate alarms that correlate multiple occurrences on the network. These events are classified accordingly to their risk. An application that allows the sonification of events generated by a Security Information Event Management can facilitate the security monitoring of a home network by a less tech-savvy user by allowing him to just listen to the result of the sonification of such events.

Abstract
The increase in usage of smartphones and the ubiquity of Internet access have made mobile communications services very attractive to users. Messaging services are among the most popular services on the Internet. In recent years, this services started to support confidentiality and anonymity. A recurrent problem with the existing messaging solutions is their lack of resistance to impersonation attacks. The proposed solution addresses the impersonation problem, without neglecting user confidentiality and anonymity, by forcing users to exchange the required cryptographic material among themselves. Moreover, this exchange must use a proximity communication technology, forcing the users to physically meet.

Abstract
The expansion of social networks has contributed to the propagation of information relevant to general audiences. However, this is small percentage compared to all the data shared in such online platforms, which also includes private/personal information, simple chat messages and the recent called ‘fake news’. In this paper, we make an exploratory analysis on two social networks to extract features that are indicators of relevant information in social network messages. Our goal is to build accurate machine learning models that are capable of detecting what is journalistically relevant. We conducted two experiments on CrowdFlower to build a solid ground truth for the models, by comparing the number of evaluations per post against the number of posts classified. The results show evidence that increasing the number of samples will result in a better performance on the relevancy classification task, even when relaxing in the number of evaluations per post. In addition, results show that there are significant correlations between the relevance of a post and its interest and whether is meaningfully for the majority of people. Finally, we achieve approximately 80% accuracy in the task of relevance detection using a small set of learning algorithms. © 2017 Copyright is held by the owner/author(s).

Abstract
Social networks are becoming a wide repository of information, some of which may be of interest for general audiences. In this study we investigate which features may be extracted from single posts propagated throughout a social network, and that are indicative of its relevance, from a journalistic perspective. We then test these features with a set of supervised learning algorithms in order to evaluate our hypothesis. The main results indicate that if a text fragment is pointed out as being interesting, meaningful for the majority of people, reliable and with a wide scope, then it is more likely to be considered as relevant. This approach also presents promising results when validated with several well-known learning algorithms.