Abstract
In this paper we present a complete system for object tracking over multiple uncalibrated cameras with or without overlapping fields of view. We employ an approach based on the bag-of-visterms technique to represent and match tracked objects. The tracks are compared with a global object model based on an ensemble of individual object models. The system can globally recognise objects and minimise common tracking problems such as track drift or split. The output is a timeline representing the objects present in a given multi-camera scene. The methods employed in the system are online and can be optimized to operate in real-time.

Abstract
The ever increasing number of platforms and languages available to software developers means that the software industry is reaching high levels of complexity. Model Driven Architecture (MDA) presents a solution to the problem of improving software development processes in this changing and complex environment. MDA driven development is based on models definition and transformation. Design patterns provide a means to reuse proven solutions during development. Identifying design patterns in the models of a MDA approach helps their understanding, but also the identification of good practices during analysis. However, when analyzing or maintaining code that has not been developed according to MDA principles, or that has been changed independently from the models, the need arises to reverse engineer the models from the code prior to patterns' identification. The approach presented herein consists in transforming source code into models, and infer design patterns from these models. Erich Gamma's cataloged patterns provide us a starting point for the pattern inference process. MapIt, the tool which implements these functionalities is described.

Abstract
In this paper we investigate the creation of an heterogeneous underwater network with static and mobile assets cooperating together in coordinated missions using acoustic links. Each underwater device combines communication, networking, and sensing capabilities, and cooperates with the other devices to accomplish a given task. The flexibility and capability of the proposed system allows to overcome the limitations of commercial solutions currently available in the market which typically focus on point to point communications. SUNSET framework has been used to provide acoustic communication and networking capabilities to AUVs, ASVs and moored systems developed by the Oceans Systems Group, at the University of Porto, in Portugal. New solutions have been developed and tested allowing to combine together acoustic data transmission and ranging estimation, to control the underwater nodes acoustically and to instruct the vehicles on keeping a given formation using acoustic links. To validate the proposed approach several experiments with increasing complexity have been conducted at the laboratory and in the field. The experimental results confirm the validity, efficiency and reliability of the proposed solution opening to several possibilities for future developments.

Abstract
"Traditional" clustering, in broad sense, aims at organizing objects into groups (clusters) whose members are "similar" among them and are "dissimilar" to objects belonging to the other groups. In contrast, in conceptual clustering the underlying structure of the data together with the description language which is available to the learner is what drives cluster formation, thus providing intelligible descriptions of the clusters, facilitating their interpretation. We present a novel conceptual clustering system for multi-relational data, based on the popular k?-?medoids algorithm. Although clustering is, generally, not straightforward to evaluate, experimental results on several applications show promising results. Clusters generated without class information agree very well with the true class labels of cluster's members. Moreover, it was possible to obtain intelligible and meaningful descriptions of the clusters. © 2012 Springer-Verlag Berlin Heidelberg.

Abstract
Nowadays, the large penetration of wind power generation poses new challenges for the dynamic voltage stability analysis of an electric power system. In this paper is studied the influence of the reactive power management for wind power plants in the dynamic voltage stability of an electric power network under a fault condition, considering two different Doubly-Fed Induction Generators (DFIG) models. The simulation results were obtained using the EUROSTAG software package.

Abstract
The development of groupware systems can be supported by the perspectives provided by taxonomies categorizing collaboration systems and theoretical approaches from the multidisciplinary field of Computer-Supported Cooperative Work (CSCW). In the last decades, multiple taxonomic schemes were developed with different classification dimensions, but only a few addressed the socio-technical perspective that encompasses the interaction between groups of people and technology in work contexts. Moreover, there is an ambiguity in the use of the categories presented in the literature. Aiming to tackle this vagueness and support the development of future groupware systems aware of social phenomena, we present a comprehensive classification model to interrelate technological requirements with CSCW dimensions of communication, coordination, cooperation, time and space, regulation, awareness, group dynamics, and complementary categories obtained from a taxonomic literature review. © 2012 Springer-Verlag.