Abstract
This study explored the viability of out-the-box, pre-trained ConvNet models as a tool to generate features for large-scale classification tasks. A juxtaposition with generative methods for vocabulary generation was drawn. Both methods were chosen in an attempt to integrate other datasets (transfer learning) and unlabelled data, respectively. Both methods were used together, studying the viability of a ConvNet model to estimate category labels of unlabelled images. All experiments pertaining to this study were carried out over a two-class set, later expanded into a 5-category dataset. The pre-trained models used were obtained from the Caffe Model Zoo. The study showed that the pre-trained model achieved best results for the binary dataset, with an accuracy of 0.945. However, for the 5-class dataset, generative vocabularies outperformed the ConvNet (0.91 vs. 0.861). Furthermore, when replacing labelled images with unlabelled ones during training, acceptable accuracy scores were obtained (as high as 0.903). Additionally, it was observed that linear kernels perform particularly well when utilized with generative models. This was especially relevant when compared to ConvNets, which require days of training even when utilizing multiple GPUs for computations.

Abstract
This work explores the application of information technologies in the processes of mentoring, allowing to optimize them and adapting this service of mentoring, to the current reality. It is intended to understand the importance of e-mentoring for the improvement and achievement of objectives for both apprentices and mentors, in which it shows how the relationship between both should be done, in order to be able to communicate with knowledge sharing throughout the relationship, from its beginning to its closure.

Abstract
Sous vide cook-chill (SVCC) is characterized by vacuum-packaging raw or partially prepared foods before pasteurization, followed by rapid chilling and storage below 3C. The application of essential oils (EOs) to food products is a suitable strategy to control pathogens and to extend their shelf life by reducing microbial levels and oxidative processes. The aim of this study was to evaluate the efficacy of Rosmarinus officinalis L. (rosemary) and Thymus vulgaris L. (thyme) EOs against Listeria monocytogenes ATCC 679, inoculated in beef processed by SVCC stored at 2 and 8C for 1, 2, 3, 7, 14, 21 and 28 days. The composition of EOs was identified by gas chromatography-mass spectrometry analysis. The main compounds identified in rosemary EO were eucalyptol (13.05%), camphor (8.93%), verbenone (8.58%), endo-borneol (7.87%) and alpha-pinene (6.78%) and in thyme EO were linalool (18.18%), thymol (7.48%), limonene (6.49%), endo-borneol (5.86%) and terpinen-4-ol (5.66%). Using the minimum inhibitory concentration (MIC) method, L. monocytogenes was inhibited at 3.9 mu L/mL to thyme EO and at 62.5 mu L/mL to rosemary EO. Beef samples of M. longissimus thoracis et lumborum were packaged in bags, inoculated and one of each EO was added at MIC values. Bags were vacuum-sealed and samples were processed at 55C/65 min for 3log(10) CFU/g reduction. A reduction of the counts of L. monocytogenes was observed in all samples at 2C. At 8C counts of L. monocytogenes were almost similar in control samples and those with thyme EO with an increase of themicrobial counts since day 7. Inversely, counts of L. monocytogenes in beef samples with rosemary EO stored at 2 and 8C decreased about 2 log(10) CFU. These results support the possibility of using rosemary EO as natural preservative due to its antimicrobial effect against L. monocytogenes. Also, our results confirm that an adequate chilling storage is essential to guarantee the safety of SVCC product regarding L. monocytogenes to avoid foodborne outbreaks.

Abstract
Collaborative working as well as sharing resources and knowledge represent key points in today's development in all fields, including education. Know-how transfer and collaboration in learning and teaching are aspects promoted and sustained by institutional management as well as the European initiatives. Thus, leading to the idea of a federation which will facilitate engineering education. A consortium formed by five European universities decided to join efforts to provide to the community a federation, which could be used by different stakeholders interest in teaching, learning or developing new skills in the field of electronics. The proposed remote system, Virtual Instruments System in Reality, or VISIR in short, offers the possibility of working with real equipment and obtain the real-world/real-time measurements. By developing such a VISIR federation some of the constraints of using remote labs, the ones associated with development and maintenance costs, and scalability, will be minimized. This paper aims to present the initial steps for developing a VISIR Federation, which is also the primary goal of PILAR Platform Integration of Laboratories based on the Architecture of visiR project.

Abstract
"Masseiras" is an ancient Portuguese agriculture system, where soil was developed from sand dunes enriched with seaweeds over more than a century. Due to the importance for the local economy, this system evolved for greenhouse structures. In this study we compared the bacterial community composition and structure of "Masseiras" soil, aiming at assessing the potential impact of different agricultural practices. The bulk soil of two greenhouses (following or not the recommended agriculture good practices, FGP and NFGP, respectively) was compared based on their physicochemical properties and bacterial community. In both FGP and NFGP, Proteobacteria, Acidobacteria, Actinobacteria, Bacteroidetes, Firmicutes, and Gemmatimonadetes were in a proportion of 5:1:1:1:1:1. However, the bacterial community of soil FGP was richer and more diverse than that of soil NFGP. Members of the classes Bacilli and Gemm-1, with higher relative abundance in NFGP and FGP, respectively, were those contributing most for distinguishing the bacterial communities of both soils. The differences in the structure of the bacterial communities correlated (Mantel test) with some soil physicochemical properties, such as electrical conductivity and nitrate and Zn contents, which were significantly higher in soil NFGP than in soil FGP.

Abstract
Social Media has been disrupting traditional technology mediated learning, providing students and educators with unsupervised and informal tools and spaces where authentic learning occurs. Still, the traditional LMS persists as the core element in this context, while lacking additional management, monitoring and analysis tools to handle informal learning and content. In this paper, we present an integrated methodology that combines social network analytics, sentiment analysis and topic categorization to perform social content visualizations and analysis aimed at integrated learning environments. Results provide insights on networked content dimension, type of structure, degree of popularity and degree of controversy, as well as on their educational and functional potential in the field of learning analytics.