Abstract
Reducing the dimensionality of datasets is a fundamental step in the task of building a classification model. Feature selection is the process of selecting a smaller subset of features from the original one in order to enhance the performance of the classification model. The problem is known to be NP-hard, and despite the existence of several algorithms there is not one that outperforms the others in all scenarios. Due to the complexity of the problem usually feature selection algorithms have to compromise the quality of their solutions in order to execute in a practicable amount of time. Parallel computing techniques emerge as a potential solution to tackle this problem. There are several approaches that already execute feature selection in parallel resorting to synchronous models. These are preferred due to their simplicity and capability to use with any feature selection algorithm. However, synchronous models implement pausing points during the execution flow, which decrease the parallel performance. In this paper, we discuss the challenges of executing feature selection algorithms in parallel using asynchronous models, and present a feature selection algorithm that favours these models. Furthermore, we present two strategies for an asynchronous parallel execution not only of our algorithm but of any other feature selection approach. The first strategy solves the problem using the distributed memory paradigm, while the second exploits the use of shared memory. We evaluate the parallel performance of our strategies using up to 32 cores. The results show near linear speedups for both strategies, with the shared memory strategy outperforming the distributed one. Additionally, we provide an example of adapting our strategies to execute the Sequential forward Search asynchronously. We further test this version versus a synchronous one. Our results revealed that, by using an asynchronous strategy, we are able to save an average of 7.5% of the execution time.

Abstract
We describe a case study application for untethered video dissemination using a hybrid edge-cloud architecture featuring Android devices, possibly organised in WiFi-Direct groups, and Raspberry Pi-based cloudlets, structured in a mesh and also working as access points. The application was tested in the real-world scenario of a Portuguese volleyball league game. During the game, users of the application recorded videos and injected them in the edge-cloud. The cloudlet servers continuously synchronised their cached video contents over the mesh network, allowing users on different locations to share their videos, without resorting to any other network infrastructure. An analysis of the logs gathered during the experiment shows that such portable setups can easily disseminate videos to tens of users through the edge-cloud with low latencies. We observe that the edge-cloud may be naturally resilient to faulty cloudlets or devices, taking advantage of video caching within devices and WiFi-Direct groups, and of device churn to opportunistically disseminate videos.

Abstract
Given a set of temporal networks, from different domains and with different sizes, how can we compare them? Can we identify evolutionary patterns that are both (i) characteristic and (ii) meaningful? We address these challenges by introducing a novel temporal and topological network fingerprint named Graphlet-orbit Transitions (GoT). We demonstrate that GoT provides very rich and interpretable network characterizations. Our work puts forward an extension of graphlets and uses the notion of orbits to encapsulate the roles of nodes in each subgraph. We build a transition matrix that keeps track of the temporal trajectory of nodes in terms of their orbits, therefore describing their evolution. We also introduce a metric (OTA) to compare two networks when considering these matrices. Our experiments show that networks representing similar systems have characteristic orbit transitions. GoT correctly groups synthetic networks pertaining to well-known graph models more accurately than competing static and dynamic state-of-the-art approaches by over 30%. Furthermore, our tests on real-world networks show that GoT produces highly interpretable results, which we use to provide insight into characteristic orbit transitions.

Abstract
Linking an expert to his knowledge areas is still a challenging research problem. The task is usually divided into two steps: identifying the knowledge areas/topics in the text corpus and assign them to the experts. Common approaches for the expert profiling task are based on the Latent Dirichlet Allocation (LDA) algorithm. As a result, they require pre-defining the number of topics to be identified which is not ideal in most cases. Furthermore, LDA generates a list of independent topics without any kind of relationship between them. Expert profiles created using this kind of flat topic lists have been reported as highly redundant and many times either too specific or too general. In this paper we propose a methodology that addresses these limitations by creating hierarchical expert profiles, where the knowledge areas of a researcher are mapped along different granularity levels, from broad areas to more specific ones. For the purpose, we explore the rich structure and semantics of Heterogeneous Information Networks (HINs). Our strategy is divided into two parts. First, we introduce a novel algorithm that can fully use the rich content of an HIN to create a topical hierarchy, by discovering overlapping communities and ranking the nodes inside each community. We then present a strategy to map the knowledge areas of an expert along all the levels of the hierarchy, exploiting the information we have about the expert to obtain an hierarchical profile of topics. To test our proposed methodology, we used a computer science bibliographical dataset to create a star-schema HIN containing publications as star-nodes and authors, keywords and ISI fields as attribute-nodes. We use heterogeneous pointwise mutual information to demonstrate the quality and coherence of our created hierarchies. Furthermore, we use manually labelled data to serve as ground truth to evaluate our hierarchical expert profiles, showcasing how our strategy is capable of building accurate profiles. © 2018, Springer Nature Switzerland AG.

Abstract
Evaluating scientists based on their scientific production is often a controversial topic. Nevertheless, bibliometrics and algorithmic approaches can assist traditional peer review in numerous tasks, such as attributing research grants, deciding scientific committees, or choosing faculty promotions. Traditional bibliometrics focus on individual measures, disregarding the whole data (i.e., the whole network). Here we put forward OTARIOS, a graph-ranking method which combines multiple publication/citation criteria to rank authors. OTARIOS divides the original network in two subnetworks, insiders and outsiders, which is an adequate representation of citation networks with missing information. We evaluate OTARIOS on a set of five real networks, each with publications in distinct areas of Computer Science. When matching a metric’s produced ranking with best papers awards received, we observe that OTARIOS is >20 more accurate than traditional bibliometrics. We obtain the best results when OTARIOS considers (i) the author’s publication volume and publication recency, (ii) how recently his work is being cited by outsiders, and (iii) how recently his work is being cited by insiders and how individual he his. © 2019, Springer Nature Switzerland AG.

Abstract