Abstract

Abstract
Given an heterogeneous social network, can we forecast its future? Can we predict who will start using a given hashtag on twitter? Can we leverage side information, such as who retweets or follows whom, to improve our membership forecasts? We present TENSORCAST, a novel method that forecasts time-evolving networks more accurately than current state of the art methods by incorporating multiple data sources in coupled tensors. TENSORCAST is (a) scalable, being linearithmic on the number of connections; (b) effective, achieving over 20% improved precision on top-1000 forecasts of community members; (c) general, being applicable to data sources with different structure. We run our method on multiple real-world networks, including DBLP and a Twitter temporal network with over 310 million non-zeros, where we predict the evolution of the activity of the use of political hashtags.

Abstract

Abstract
Location-based games require, among other things, physical activity and real-world context. Additionally, ensuring that the players are assigned challenges that are adequate and safe for the current context (both physical and spatial) is also important, as it can improve both the gaming experience and the outcomes of the exercise. However, the impact adaptivity has in the specific case of location-based exergames still has not been researched in depth. In this paper, we present a location-based exergame capable of adapting its mechanics to the current context.

Abstract
In the vehicle routing problem with heterogeneous fleet and variable costs (HFVRP), the group of routes to be developed to satisfy the demand of the customer must be determined, considering the minimization of the total costs of the travelled distance. Heuristic algorithms based on local searches use simple movements (neighborhoods) to generate feasible solutions to problems related to route design. In this article, we conduct a comparative analysis of granular neighborhoods in a Tabu Search for the HFVRP, in terms of the quality of the obtained solution. The computational experiments, performed on instances of benchmarking for the HFVRP, showed the efficiency and effectiveness of implementing some neighborhoods in metaheuristic algorithms of path, such as the Tabu Search.

Abstract
This paper aims at presenting the STRONGMAR Summer School 2016 that took place at the Base Naval de Lisboa, of the Portuguese Navy. The STRONGMAR project ideal motivates the development of maritime and marine science research and technology through the knowledge transfer between INESC TEC and promising, and prestigious, leading research European institutions. This process takes place through theoretical lectures and training, and via experimental application of the concepts discussed in order to further develop technology related to the sea environment. The practical application of the STRONGMAR project ideal takes place during events such as summer schools, winter schools, thematic workshops and scientific conferences. The STRONGMAR Summer School 2016 approaches the subject of "Introduction to Advanced Marine Technologies", providing a strong component of practical applications in underwater archaeology. It develops the study of unmanned systems applied to underwater archaeology, through the use of unmanned underwater vehicles. As a whole, this paper describes the Summer School experience, providing some results and greater insight on the topic of underwater archaeology.