Publicaciones

Technology played a crucial role in combating the COVID pandemic, both in the rapid development of vaccines and the early detection of the virus. Consequently, numerous studies in the medical field have focused on leveraging the power of artificial intelligence for COVID-19 detection. However, in the medical domain, it is essential to have a clear understanding of the processes and algorithms used in decision-making, as these directly impact people’s health. Therefore, efforts have been made to implement explainable artificial intelligence techniques, enabling humans to understand and explain the deep learning algorithms used in disease detection. In this work, we present an approach to detecting COVID-19 in chest X-rays that combines reconstruction-based anomaly discov- ery with perturbation-based attribution. Specifically, we use a Variational Autoencoder trained on healthy lungs to identify lung anomalies, and we additionally evaluate a Per- ceptual Autoencoder (PAE) that incorporates a perceptual loss to produce sharper recon- structions and more contrastive residual maps. These signals are then used to highlight critical image evidence through both patch-level scoring (grid-based occlusion) and pixel- level masking derived from reconstruction-error maps, enabling healthcare professionals to localize relevant regions more effectively. Moreover, the proposed framework provides clearer explanations of the model’s decisions by quantifying the prediction change after occluding the identified regions, reducing the complexity of the “black boxes” generated by deep learning neural networks. With this methodology, we aim to improve the effec- tiveness and reliability of early COVID-19 detection through chest X-rays.

La inteligencia artificial (IA) generativa ha introducido transformaciones significativas en la docencia universitaria, generando oportunidades para el aprendizaje personalizado, el apoyo cognitivo y la producción académica. Sin embargo, su adopción responsable sigue enfrentándose a desafíos por los persistentes riesgos éticos, las brechas de competencias y la falta de marcos pedagógicos y regulatorios consolidados. Este artículo presenta los resultados de un proyecto de innovación educativa desarrollado en la Universidad Católica de Temuco, orientado a explorar y validar usos pedagógicos de la IA mediante un enfoque interdisciplinar en las facultades de Ciencias Jurídicas, Ciencias Básicas e Ingeniería. Empleando una metodología de investigación-acción y un diseño de método mixto basado en encuestas pre y posintervención, se evidenció un aumento significativo del uso académico de herramientas de IA y una percepción positiva sobre su impacto en el aprendizaje, lo que destacó que un 68,9% del estudiantado identificó mejoras en su trayectoria formativa. No obstante, un 41,34% mantiene una valoración ética cautelosa, especialmente en contextos jurídicos. Los resultados subrayan la necesidad de superar una visión instrumental de la IA hacia enfoques centrados en el proceso de aprendizaje, el fortalecimiento de competencias transversales y la reafirmación del rol docente como mediador crítico del uso tecnológico.

Demand for image inspection in dermatology is growing, leading to the exploration of deep learning techniques and algorithms to enhance diagnostic accuracy and reduce healthcare service times. Our approach includes employing artificial techniques to improve image quality for more effective model outcomes. Diagnosing skin diseases is visually difficult without dermatological expertise, as some skin lesions may result from other diseases. We utilized a public database for testing image enhancement which contains direct skin disease labels and available patient metadata (when present) for nuanced model predictions. Enhancement methods like ExcNet, CLAHE, and median filter addressed issues such as poor framing and inadequate lighting. The models tested included advanced neural networks such as CNN, ResNet-50, and MobileNet, among others. In our setting, low-light conditions were simulated via OpenCV darkening, and results should be interpreted accordingly. Within these constraints, the ExcNet deep learning technique showed the highest validation/test performance, with test accuracy reaching up to 91% in our experiments, while acknowledging potential label noise and dataset scope limitations.

Technology played a crucial role in combating the COVID pandemic, both in the rapid development of vaccines and the early detection of the virus. Consequently, numerous studies in the medical field have focused on leveraging the power of artificial intelligence for COVID-19 detection. However, in the medical domain, it is essential to have a clear understanding of the processes and algorithms used in decision-making, as these directly impact people’s health. Therefore, efforts have been made to implement explainable artificial intelligence techniques, enabling humans to understand and explain the deep learning algorithms used in disease detection. In this work, we present a novel approach to detecting COVID-19 in chest X-rays using a Variational Autoencoder model to identify lung anomalies. This methodology aims to highlight critical areas of the image, allowing healthcare professionals to identify them more effectively. Additionally, it seeks to provide clearer explanations of the decisions made by the artificial i ntelligence, r educing the complexity of the "black boxes" generated by deep learning neural networks. With this methodology, we hope to improve the effectiveness and reliability of early COVID-19 detection through chest X-rays.

Gating-modifying peptide toxins preferentially bind to the voltage-sensing domain KV4.3 voltage-gated potassium channels, and the molecular determinant of this interaction remains unclear. Through unconstrained multiscale molecular dynamics simulations, we could recapitulate spontaneous binding of a gating-modifying toxin, such as SNX-482 (from tarantula Hysterocrates gigas) to the KV4.3 potassium channel at the membrane interface, overcoming the limitations of traditional docking methods. This approach revealed two likely binding poses centered on the S3-S4 linker of KV4.3, one of which included pore region residues. By replacing residues in the modeled binding poses and quantifying SNX-482 (1 μM) effects on the voltage dependence of activation of KV4.3 channels expressed in Xenopus oocytes, we determined that only the S3-S4 linker is necessary to retain SNX-482 binding. Notably, we identified M276 from the S3-S4 linker as a key stabilizing element of SNX-482/KV4.3 complex, and M276A substitution ablates toxin sensitivity. Overall, our study establishes a framework for rational drug design whose target sites lie at the membrane interface, such as gating modifiers of A-type potassium channels, relevant to neurological and cardiovascular diseases.

Like much of Latin America, Chile’s urban and rural areas present marked contrasts in population, production, and environment. The contribution of this article is to define rural typologies based on the census districts of Chile and to use 35 spatial variables for cluster analysis by Affinity Propagation (AP). AP is a clustering technique that works through the transmission of ‘messages’ between data points, seeking to represent the similarity between these points and find clusters adaptively, thus contributing to the generation of information on common patterns and characteristics in rural territories, which facilitates decision making for territorial planning and development. The results made it possible to identify 95 clusters, which were subsequently reclassified into 34 macroclusters and six macrozones. This analysis contributes to identify the particularities of rural territories with a spatial connotation, providing information for the elaboration of future detailed diagnoses that will serve as a basis for the design of new rural policies.

High voltage-activated (HVA) calcium channels (CaV) have four homologous but nonidentical repeats encompassing a voltage-sensing domain (VSD) and a quarter of the pore domain (PD). HVA can be modulated by at least two accessory subunits α2δ and CaVβ. A long-standing issue is how cytoplasmic CaVβ can shift the voltage dependence of channel opening without altering gating currents. Tracking the movement of individual VSDs by voltage-clamp fluorometry in human CaV1.2 revealed that only the VSD from the second repeat (VSD II) is perturbed by CaVβ3 in a construct combining a fluorophore-tagged VSD II (S1623C) with a quenching tryptophan within 11 Å in the PD of repeat III (E1141W). The final construct, S612C_E1141W, exhibited a biphasic voltage-dependent fluorescence whose negative phase was enhanced by CaVβ3. This behavior was well described by a kinetic model that includes three states for VSD II of which the intermediate state contributes the most to pore opening in a CaVβ-dependent manner, and that open channels with VSD II in the intermediate state would yield the lowest fluorescence emissions. Molecular dynamics simulation correlates a structure with two translocated arginines with frequent fluorophore-W contact between VSD II and the pore of open channels.

Pelagic fish have evolved specialized biogenic multilayer reflectors composed of stacks of intracellular anhydrous guanine crystals separated by cytoplasm, giving notorious silvery appearance to their skin. While the reflective properties of guanine crystals and their utility for fish camouflage have been shown in other fish species, this is the first evaluation on fish species from the southern hemisphere, and from the Humboldt current system. This is one of the most productive systems on earth, having particular oceanographic conditions such as upwelling, and thus under strong selection pressures. In this study, we conducted a comparative analysis of four pelagic species, Sardine, Anchovy, and Snoek, known for their silvery characteristics, and Mote sculpin, which lacks silvery features. We aimed to explore the biological mechanisms underlying light reflectivity in fish species and to understand how fish skin microstructures affect whole fish light reflectance and intensity in the visible spectrum. We measured the reflectance of individual fish using hyperspectral imaging and characterized the guanine crystal/cytoplasm layers within the skin of each fish using high-resolution scanning electron microscopy. These Scanning Electron Microscopy (SEM) images were analyzed using the 2D discrete Fourier transform to extract the spatial patterns that govern the light interaction with the guanine crystal structures. A novel spatial frequency analysis approach applied to SEM images explained reflectance differences between species with similar spectral behavior. Furthermore, this study presents the first fish classifiers based on the analysis of spatial frequency features, achieving up to 92.14% accuracy using a K-Nearest Neighbors classifier, highlighting the functional and taxonomic relevance of guanine microstructure organization. Our findings confirm, on pelagic fish species from the Humboldt current system, that silvery species have a chaotic distribution/arrangement of guanine crystals, whereas non-silvery species have a more organized arrangement. Accordingly, Fourier analysis indicated that silvery fish are capable of scattering light uniformly across the visible spectrum. In contrast, the Mote sculpin shows a stronger scattering of red light, distinguishing it from silvery fish.

Background/Objectives: Bcr-Abl inhibitors such as imatinib have been used to treat chronic myeloid leukemia (CML). However, the efficacy of these drugs has diminished due to mutations in the kinase domain, notably the T315I mutation. Therefore, in this study, new purine derivatives were designed as Bcr-Abl inhibitors based on 3D-QSAR studies. Methods: A database of 58 purines that inhibit Bcr-Abl was used to construct 3D-QSAR models. Using chemical information from these models, a small group of new purines was designed, synthesized, and evaluated in Bcr-Abl. Viability assays were conducted on imatinib-sensitive CML cells (K562 and KCL22) and imatinib-resistant cells (KCL22-B8). In silico analyses were performed to confirm the results. Results: Seven purines were easily synthesized (7a–g). Compounds 7a and 7c demonstrated the highest inhibition activity on Bcr-Abl (IC50 = 0.13 and 0.19 μM), surpassing the potency of imatinib (IC50 = 0.33 μM). 7c exhibited the highest potency, with GI50 = 0.30 μM on K562 cells and 1.54 μM on KCL22 cells. The GI50 values obtained for non-neoplastic HEK293T cells indicated that 7c was less toxic than imatinib. Interestingly, KCL22-B8 cells (expressing Bcr-AblT315I) showed greater sensitivity to 7e and 7f than to imatinib (GI50 = 13.80 and 15.43 vs. >20 μM, respectively). In silico analyses, including docking and molecular dynamics studies of Bcr-AblT315I, were conducted to elucidate the enhanced potency of 7e and 7f. Thus, this study provides in silico models to identify novel inhibitors that target a kinase of significance in CML.

When a company decides to use a chatbot, it must evaluate enough items to be able to take the right decision as to which one you should use to satisfy all your needs. There are times in which this decision is biased by trying to use the latest technologies available in the market and with this the full potential of the available products is not used. This project seeks to carry out research to implement an interaction chatbot basic text that allows it to be scalable mixed with other technologies such as ChatGPT and RPA (robotic process automation) can enhance its use to provide answers on unlimited topics added to having the possibility of carrying out tasks through RPA, which is all technology-oriented to the use of software with the aim of reducing human intervention in the use of applications computing, especially in repetitive tasks that vary very little in each iteration. RPA is a relatively new term, but not the concept behind it, since for decades it has been have looked for ways to optimize resources through software that performs specific functions in record time.

Skin cancer is a growing global concern, with cases steadily rising. Typically, malignant moles are identified through visual inspection, using dermatoscopy and patient history. Active thermography has emerged as an effective method to distinguish between malignant and benign lesions. Our previous research showed that spatio-temporal features can be extracted from suspicious lesions to accurately determine malignancy, which was applied in a distance-based classifier. In this study, we build on that foundation by introducing a set of novel spatial and temporal features that enhance classification accuracy and can be integrated into any machine learning approach. These features were implemented in a support-vector machine classifier to detect malignancy. Notably, our method addresses a common limitation in existing approaches-manual lesion selection-by automating the process using a U-Net convolutional neural network. We validated our system by comparing U-Net's performance with expert dermatologist segmentations, achieving a 17% improvement in the Jaccard index over a semi-automatic algorithm. The detection algorithm relies on accurate lesion segmentation, and its performance was evaluated across four segmentation techniques. At an 85% sensitivity threshold, expert segmentation provided the highest specificity at 87.62%, while non-expert and U-Net segmentations achieved comparable results of 69.63% and 68.80%, respectively. Semi-automatic segmentation lagged behind at 64.45%. This automated detection system performs comparably to high-accuracy methods while offering a more standardized and efficient solution. The proposed automatic system achieves 3% higher accuracy compared to the ResNet152V2 network when processing low-quality images obtained in a clinical setting.

The integration of Information and Communication Technologies (ICT) in higher education has seen continuous growth, offering numerous tools and platforms to enhance the learning process for both students and educators. However, the lack of pedagogical training among many teachers prevents the use of ICT from automatically guaranteeing quality education. This article introduces a collaborative platform designed to facilitate educators in creating learning modules. The platform considers three key elements when designing a module: i) assessment as the guiding axis for the design, ii) the SOLO taxonomy to guide the definition of the learning outcomes, and iii) a conversational framework to assist in designing the various tasks that students must complete. Although the platform showed positive results a few years ago, evaluation revealed several areas for improvement in terms of technical, pedagogical, and usability aspects. As a result, improvements were implemented on three levels. First, the user experience was optimized through significant enhancements in the interface and collaboration functionalities, making the platform more intuitive and accessible. Second, the pedagogical support was updated, replacing Bloom's taxonomy with the SOLO taxonomy, which provides a more comprehensive and detailed approach to assessing student understanding and learning outcomes. Finally, the technological implementation was enhanced using advanced software tools, allowing for more efficient and robust development of features. These improvements not only optimized the user experience but also provided stronger pedagogical support and more advanced technological implementation, ensuring that the platform remains a valuable and effective tool in higher education. This experience yielded satisfactory results, both in evaluating the platform's usage by educators and in assessing students' experiences when working with learning modules generated by the platform.

In this study, we evaluate a CUDA-based simulated annealing algorithm for solving the centralized studentto- school assignment problem. Addressing this issue is critical for enhancing educational equity and efficiency. Optimizing this assignment process not only improves logistical outcomes but also potentially reduces socioeconomic disparities within educational districts. Our objective function aims to minimize a linear combination of the mean student-school distance, which impacts travel time; district segregation, measured by the Dissimilarity Index, which reflects the equity of school placements; and the costs associated with under or overfilled schools, which affect resource allocation. Initially, the system assigns each school to a GPU block, exploring a set number of students per block defined by thread count. Each thread reassigns a student to the school designated for its block. Following this, a parallel reduction process captures the best solution at each iteration, first by threads then by blocks (many-to-many neighbor selection method). Increasing exploration through more blocks and threads accelerates the algorithm's cooling, degrading solution quality. Conversely, reducing the number of exploring schools improves outcomes, scaling positively with thread count. We thus optimize by comparing results using the same school across multiple blocks with diverse student explorations through assigned threads (one-to-many neighbor selection method). Performance, in terms of minimizing the objective function, enhances with the total number of blocks multiplied by threads. A key finding is that while a parallelized simulated annealing algorithm offers a robust tool for this problem, ambitious exploration strategies must be managed carefully to avoid falling into local minima, a paradoxical result of the algorithm's rapid cooling dynamics.

Cloud computing, typically provided by large infrastructure providers, allows companies to utilize services efficiently under the premise of cost reduction. Known as the “public cloud,” this model contrasts with the “private cloud,” often perceived as costly and complex to implement. However, when high availability is required, the complexities of both models align, and the costs associated with public cloud services can become prohibitive. This paper explores the technical feasibility and economic viability of deploying private cloud infrastructure using refurbished hardware. It presents a detailed architecture incorporating Canonical MaaS for Metal-as-a-Service (MaaS), OpenNebula for Infrastructure-as-a-Service (IaaS), and Incus for Platform-as-a-Service (PaaS), with Moodle as the application layer case study. The study finds that private clouds, especially those utilizing refurbished hardware, not only offer a costeffective alternative to public clouds but also enhance security with complete control over data and systems, provide customizable self-management to meet specific organizational needs, utilize existing resources more efficiently for greater sustainability, and ensure technological sovereignty by reducing dependency on public cloud providers. These benefits collectively underscore the strategic advantages of adopting private cloud infrastructures in certain contexts.

A novel method, to our knowledge, for monitoring weightlifting exercises based on infrared imaging is proposed in this work. For the infrared workout weightlifting recorded scenes, radiometry and artificial intelligence were employed for in-scene temperature and biomechanical athletes’ body parts position mapping. Our method was effective in monitoring muscle exertion during high-performance athletic exercises, as evidenced by the results obtained from real athletic datasets. The method generates a color-labeled sequence of thermal images and reports on body part positions, which can be used by judges and trainers to guide athletes toward safer and more efficient practices.

Recognizing variable stars is a task of interest in the astronomy community. Currently, this task has taken advantage of deep learning algorithms. However, these algorithms require a large amount of data to achieve high levels of precision. In this work, self-supervised learning is proposed to improve the classification of variable stars considering a reduced amount of data using recurrent networks. The experiments in Gaia dataset show that the proposed approach allows to improve performance, when compared with traditional initialization schemes, up to 7% and 13% in real databases in semi-supervised learning scenarios. In future work, we propose considering experiments with other variable star databases.

Infrared thermography is considered a useful technique for diagnosing several skin pathologies but it has not been widely adopted mainly due to its high cost. Here, we investigate the feasibility of using low-cost infrared cameras with microbolometer technology for detecting skin cancer. For this purpose, we collected infrared data from volunteer subjects using a high-cost/high-quality infrared camera. We propose a degradation model to assess the use of lower-cost imagers in such a task. The degradation model was validated by mimicking video acquisition with the low-cost cameras, using data originally captured with a medium-cost camera. The outcome of the proposed model was then compared with the infrared video obtained with actual cameras, achieving an average Pearson correlation coefficient of more than 0.9271. Therefore, the model successfully transfers the behavior of cameras with poorer characteristics to videos acquired with higher-quality cameras. Using the proposed model, we simulated the acquisition of patient data with three different lower-cost cameras, namely, Xenics Gobi-640, Opgal Therm-App, and Seek Thermal CompactPRO. The degraded data were used to evaluate the performance of a skin cancer detection algorithm. The Xenics and Opgal cameras achieved accuracies of 84.33% and 84.20%, respectively, and sensitivities of 83.03% and 83.23%, respectively. These values closely matched those from the non-degraded data, indicating that employing these lower-cost cameras is appropriate for skin cancer detection. The Seek camera achieved an accuracy of 82.13% and a sensitivity of 79.77%. Based on these results, we conclude that this camera is appropriate for less critical applications.

In this article, we present the R package mutualinf for computing and decomposing the mutual information index of segregation by means of recursion and parallelization techniques. The mutual information index is the only multigroup index of segregation that satisfies strong decomposability properties, both for organizational units and groups. The mutualinf package contributes by (1) implementing the decomposition of the mutual information index into a "between" and a "within" term; (2) computing, in a single call, a chain of decompositions that involve one "between" term and several "within" terms; (3) providing the contributions of the variables that define the groups or the organizational units to the overall segregation; and (4) providing the demographic weights and local indexes employed in the computation of the "within" term. We illustrate the use of mutualinf using Chilean school enrollment data. With these data, we study socioeconomic and ethnic segregation in schools.

Skin cancer is one of the most common types of cancer, whose number of cases is constantly increasing. The most used method to detect skin cancer is the biopsy. It is relevant to reduce the number of biopsies, since it is an invasive and expensive procedure, and it has limited availability in some locations. Among the most successful approaches that aim to improve skin cancer detection are the algorithms that process active infrared thermography.Here, a skin cancer detection scheme is proposed, which extracts key features from active thermography videos, and uses them in the following five classifiers: K-Nearest Neighbors, Decision tree, Random forest, Support vector machine (SVM) and eXtreme Gradient Boosting (XGBoost). Under a minimization error design criteria, the best result was performed by a SVM classifier, reaching 84.14% of accuracy and 78.92% of precision. Modifying the classifiers to ensure that all the malignant cases are detected, the best performance was also achieved by the SVM classifier, with 72.85% of accuracy and 63.95% of presicion.The proposed scheme is 15% less accurate than the best detection algorithm. However, it is easier to implement and deploy and provides a framework with key preprocessing aspects to address this detection problem using active thermography. As future work, a further exploration of features will be carried out, with the aim of improving the performance of the classifier.

Finding the minimum winning coalition (MWC) is a particular case of clustering problems constrained on the clusters’ size. As such, this problem is NP-Hard, posing an exciting challenge in optimization. In this work, we present a new adhoc algorithm to solve the MWC problem, which identifies the same MWC found by a genetic algorithm implemented under the same platform. The algorithm works deterministically and achieves a solution in tenths of a second for a real problem with high complexity, the political spectrum of the House of the 75° Congress of the USA as calculated with DW-Nominate. Hence, the solution time is two orders of magnitude better than the average time it takes for the genetic algorithm to find it.

Chile has a high natural disaster risk index and is highly vulnerable to climate change. Because of this, it is crucial to achieve widespread community awareness of disasters, in order to enhance community resilience. To address this, a solution based on web technology to allow actors from the response system, and general users later, to visualize and contrast past and present images of an area affected by a disaster have been built and explored. The results indicate that this technology can be successfully developed to meet the functional and usability requirements of software engineering professionals, as well as experts from the Chilean disaster management agency. The main conclusion from this research effort is that this proof of concept technology has the potential to effectively enhance the risk perception of communities.

Interaction spaces influence emotional regulation, due to the link between bodies, emotions and places. On the one hand, a series of emotions are projected on a certain space, on the other hand, spaces have the capacity to generate different emotional responses in people. The objective of this research was to identify the direct and indirect effects of the organization of school spaces on emotional regulation and learning. Eleven key informants participated, divided into professionals who work or have demonstrated their experience in the school educational context. The design was non-probabilistic and an analysis of the level of incidence in first and second order was carried out, in order to find the forgotten effects in the explanation of the object of study. The results show that the main indirect effects are that of the school infrastructure both on the development of respect for others and on the ability to control negative emotions, both mediated by the learning environment. On the other hand, the use of the Mapuzungun language in the school educational context has indirect effects on the behavior and regulation of the emotions of the students, having as main mediator the interpersonal relationships that occur in the school educational context.

Marketing is essential for the success of any company today, both to project itself abroad and to achieve its business objectives. The personalized approach to marketing plays a crucial role, enabling more effective interaction with potential customers and reducing errors in promotional campaigns. One way to improve personalized marketing is by using artificial intelligence to predict the effectiveness of campaigns in converting users to customers. In this study, the use of Siamese neural networks for elevation modelling in a Chilean online retail company is proposed. The results obtained using this model are presented, as well as the classical elevation modelling techniques. These results show that the Siamese neural network model achieves an increase of more than 30% compared to the area under the Qini curve, while also being competitive in other metrics. As future work, it is planned to obtain more real data to carry out a better validation of this proposal as well as to compare the model with other proposed neural networks.

Glycogen synthase kinase 3 (GSK-3) is involved in different diseases, such as manic-depressive illness, Alzheimer’s disease and cancer. Studies have shown that insulin inhibits GSK-3 to keep glycogen synthase active. Inhibiting GSK-3 may have an indirect pro-insulin effect by favouring glycogen synthesis. Therefore, the development of GSK-3 inhibitors can be a useful alternative for the treatment of type II diabetes. Aminopyrimidine derivatives already proved to be interesting GSK-3 inhibitors. In the current study, comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) have been performed on a series of 122 aminopyrimidine derivatives in order to generate a robust model for the rational design of new compounds with promising antidiabetic activity. The q2 values obtained for the best CoMFA and CoMSIA models have been 0.563 and 0.598, respectively. In addition, the r2 values have been 0.823 and 0.925 for CoMFA and CoMSIA, respectively. The models were statistically validated, and from the contour maps analysis, a proposal of 10 new compounds has been generated, with predicted pIC50 higher than 9. The final contribution of our work is that: (a) we provide an extensive structure–activity relationship for GSK-3 inhibitory pyrimidines; and (b) these models may speed up the discovery of GSK-3 inhibitors based on the aminopyrimidine scaffold. Finally, we carried out docking and molecular dynamics studies of the two best candidates, which were shown to establish halogen-bond interactions with the enzyme.Communicated by Ramaswamy H. Sarma

In the last decade, a large amount of data from vehicle location sensors has been generated due to the massification of GPS systems to track them. This is because these sensors usually include multiple variables such as position, speed, angular position of the vehicle, etc., and, furthermore, they are also usually recorded in very short time intervals. On the other hand, routes are often generated so that they do not correspond to reality, due to artifacts such as buildings, bridges, or sensor failures and where, due to the large amount of data, visual analysis of human expert is unable to detect genuinely anomalous routes. The presence of such abnormalities can lead to faulty sensors being detected which may allow sensor replacement to reliably track the vehicle. However, given the reliability of the available sensors, there are very few examples of such anomalies, which can make it difficult to apply supervised learning techniques. In this work we propose the use of unsupervised deep neural network models based on stacked autoencoders to detect anomalous routes in vehicles within Santiago de Chile. The results show that the proposed model is capable of effectively detecting anomalous paths in real data considering validation given by an expert user, reaching a performance of 82.1% on average. As future work, we propose to incorporate the use of Long Short-Term Memory (LSTM) and attention-based networks in order to improve the detection of anomalous trajectories.

The purpose of this paper is to identify the emergence of indirect trade flows prompted by the export interaction of the world’s economies. Using data on exports from the United Nations Conference on Trade and Development (UNCTAD) for the period 2016–2021, we construct an international trade network which is analyzed through the “forgotten effects theory” that identifies tuples of countries with an origin, intermediary countries, and a destination. This approach intends to spotlight something beyond the analysis of the direct trade network by the identification of second and third-order paths. The analysis using both network analyses, as well as the forgotten effect approaches, which show that the international trade network presents a hub-and-spoke behavior in contrast to most extant research finding a core-periphery structure. The structure is then comprised of three almost separated trade networks and a hub country that bridges commerce between those networks. The contribution of this article is to move the analysis forward from other works that utilize trade networks, including those of econometric nature—such as the ones based on gravity models—by incorporating indirect relationships between countries, which could provide distinctive and novel insights into the study of economic networks.

Customer service is currently of great importance because this process supports the management of customer queries, requests and disagreements. In customer service, noncompliance with resolution times as well as closing times appearsas a recurring problem, which has a direct relationship with the customer experience. An alternative to improve this process isthrough the use of natural language processing algorithms to facilitate the complaint of customers considering platforms and solutions already applied in the market. In this paper, a customer complaint classifier is proposed using Google's Dialog Flow system in a Chilean company considering the implementation of a prototype composed of a text classification model and a system that manages customer interactions. This system allows an automatic categorization of customer interactions, allowing the tasks of a customer service representative to be reduced, minimizing the risks of misclassified interactions and avoiding re-categorizations that increase resolutiontimes. The proposed model allows improvements over 50% accuracy over the classification given by the pre-existing system based on user feedback. As future work we propose to create a chatbot system that allows improving user feedback.

Currently, air pollution is a highly important issue in society due to its harmful effects on human health and the environment. The prediction of pollutant concentrations in Santiago de Chile is typically based on statistical methods or classical neural networks. Existing methods often assume that historical values are known at a fixed geographic point, such that air pollution can be predicted at a future hour using time series analysis. However, these methods are inapplicable when it is necessary to know the pollutant concentrations at every point of the space. This work proposes a method that addresses the space-time prediction of PM2.5 concentration in Santiago de Chile at any spatial points through the use of the LSTM recurrent network model. In particular, by considering historical values of air pollutants (PM2.5, PM10 and nitrogen dioxide) and meteorological variables (temperature, wind speed and direction and relative humidity), measured at fixed monitoring stations, the proposed model can predict PM2.5 concentrations for the next 24 h in a new location where measurements are not available. This work describes the experiments carried out, with particular emphasis on the pre-processing step, which constitutes an important factor for obtaining relatively good results. The proposed multilayer LSTM model obtained R2 values equal to 0.74 and 0.38 in seven stations when considering forecasts of 1 and 24 h, respectively. As future work, we plan to include more input variables in the proposed model and to use attention-based networks.

Currently, cybersecurity has become one of the most important issues in the society, due to the growing technological evolution and use of digital platforms by organizations to connect with their users. A non-invasive biometric way of accessing an organizational platform is by verifying the presence of the user's face given by selfie photography in a database of authorized users. However, this procedure requires the prior construction of a database of authorized users, which prevents online verification of a person’s identity. A feasible way to carry out an online verification is by comparing the photos of the identity document and the person’s selfie. A system that allows this verification will improve the security of online use of a platform that prevents fraud and identity theft. In this work, we propose a deep neural network that allows verifying the identity of a person considering as inputs the photographs of an identity document and a selfie. This document presents experiments with various neural networks considering a public database of real camera photographs and identity documents. The results show that the use of a deep neural network with an ArcFace loss function configured with the database images of ID photos achieved a recognition rate of over 94%. Additionally, we test this procedure in a small validation sample of Chilean people obtaining similar rate. As future work, we propose the use of larger databases based on Chilean document data.

Spider toxin SNX-482 is a cysteine-rich peptide that interferes with calcium channel activity by binding to voltage-sensing domains of the CaV2.3 subtype. Two mechanisms dominate the binding process of cysteine-rich peptides: direct binding from the aqueous phase or through lateral diffusion from the membrane, the so-called reduction in dimensionality mechanism. In this work, via coarse-grained and atomistic molecular dynamics simulations, we have systematically studied the spontaneous partitioning of SNX-482 with membranes of different anionic compositions and explored via diffusional analysis both binding mechanisms. Our simulations revealed a conserved protein patch that inserts in the membrane, a preference for binding towards partially negatively charged membranes, and that electrostatics guides membrane binding by incrementing and aligning the molecular dipole. Finally, diffusivity calculations showed that the toxin diffusion along the membrane plane is an order of magnitude slower than the aqueous phase suggesting that the critical factor in determining the SNX-482-CaV2.3 binding mechanism is the affinity between the membrane and SNX-482.

ResumenEl principal objetivo de este estudio es determinar el impacto de la participación femenina en la producción científica.Se utilizan datos publicados por el Consejo Nacional de Educación (CNED, 2020) y se examina la producción científica de 63 universidades chilenas en el período 2009-2019.Se analiza la distribución por género y se utiliza el modelo econométrico de efectos fijos con datos de panel.Los resultados muestran que a nivel de universidades persiste la desigualdad según sexo en la matrícula estudiantil y en la planta académica, la cual incide negativamente en la productividad científica.Se concluye que en la medida en la que las universidades desarrollen políticas de equidad en la participación de la mujer a nivel de matrícula estudiantil o en su planta académica se debería apreciar un aumento en la productividad científica.

ASTM D6433 is used to assess the need for maintenance of pavement sections. Although the Pavement Condition Index (PCI) factor calculation method provides reliable values, this method analyzes sections and defects individually and indicates current maintenance needs, but it cannot be used to predict the occurrence of new defects. Therefore, it is necessary to complement this method by considering variables that influence the occurrence of faults, among which are the geospatial distribution and the specific characteristics of the slabs. This research focuses on the identification of multiple types of disturbances that exist in Portland Cement Pavements (PCC), located in a high traffic area in the city of Valdivia (Chile). A spatial geostatistical relationship is established through visual inspection using geographical maps, as well as distribution, using the kriging method. This technique makes use of variograms that allow quantifying the parameters used in this study, thus expressing the spatial autocorrelation of the faults analyzed. From the results obtained by spatial geostatistics and kriging, it is possible to generate a data correlation for the distribution and characteristics of the streets considered. In addition, a co-kriging method is established instead of an ordinary kriging method. The relationship between observed and predicted values improved from 0.3327 to 0.5770. The width of the slabs, as well as some streets, is shown in our analysis to be unimportant. For better model accuracy, the number of covariates associated with the type of vehicle traffic, the age and shape of the slabs, and the construction techniques used for the pavement needs to increase.

Al declarar que el Objetivo de Desarrollo Sostenible 9 activa otros de los diecisiete objetivos propuestos en 2015 por Naciones Unidas en esta materia, la Organización Mundial de la Propiedad Intelectual induce a que los gobiernos incentiven la relación entre la universidad y la industria, como principales productoras de resultados de investigación patentables. A partir de datos recogidos entre 2017 y 2018 en las siete universidades que más patentan en Chile, encontramos que, primero, existen relaciones entre los entrevistados y la industria; segundo, la mayoría de los entrevistados ha patentado; tercero, estas patentes no surgen de proyectos industria-universidad; y cuarto, la patente no se percibe como un incentivo fundamental para el desarrollo. Por tanto, el modelo chileno de producción de investigación y desarrollo (I+D) parece más lineal que de triple hélice.

Since the beginning of 2020, the diagnosis of the COVID-19 virus has been a major problem that has affected the lives of millions of people around the world. The detection time for COVID-19 with a standard detection method ranges from approximately 1 to 5 days. An efficient and fast way to detect the presence of both the COVID-19 virus is through the use of artificial intelligence (AI) techniques applied to images obtained by lung radiography. Typically, AI algorithms to detect COVID-19 consider the whole picture. However, there may be parts that affect the performance of the classifier. Furthermore, these algorithms do not indicate which is the most relevant area of this disease. In this work, we propose a deep learning approach to detect the presence of COVID-19 in lung images by recognizing the most relevant areas affected by the virus without considering human supervision. In the experiment, we considered different proposals, where the best one obtained an 88% reduction of the logit loss with respect to the baseline based on random regions near the center of the image.

Marketing is a strategy that every company must implement today within its global plan both due to the need for external projection and for the achievement of commercial objectives. Currently, personalized marketing is key to the development of a company since it allows a better interaction with potential customers and the margin of loss or error in the direction of promotional campaigns is greatly reduced. One possibility to improve personalized marketing is the prediction of the effectiveness of campaigns to transform users into customers using artificial intelligence, so it is necessary to develop models that allow identifying profiles or segments of people who are more willing to answer positively to a campaign. This task corresponds to uplift modeling that predicts the incremental impact of the application of treatments on a population, which is typically performed using classical models such as the one-model approach, the class transformation approach, and the two-model approach. In this work, the use of multilayer neural networks is proposed to perform uplift modeling in a Chilean online retail company. The results of the proposed model as well as classical uplift modeling techniques are presented. These results indicate that the neuronal model allows an increase of more than 30 % in relation to the area of the Qini curve, while it is competitive in other metrics. As future work, it is planned to model a Siamese neural network with the cost function of uplift modeling directly.

We propose two ordinal classifiers (logistic regression and LightGBM) to estimate the school learning level (obtained by an ordinal transformation of the school’s average score in the math section of the SIMCE test) for students in the 4 th grade of primary schooling. We trained these classifiers with socioeconomic variables that characterize Chilean schools. This dataset included Alonso-Villar and Del Río’s local segregation index to measure socioeconomic and gender segregation at the school level compared to the district level. For the socioeconomic segregation, we used a vulnerability criterion based on the status of a student as a recipient of the subsidy established by Law 20.248 (SEP). We used a greedy algorithm based on Variance Inflation Factor (VIF) and the Prediction Power Score (PPS) to automatically select the most orthogonal features in this dataset. This algorithm selected six variables, being the most important the self-esteem and motivation index, and both the socioeconomic and gender segregation of the school. Due to the inherent imbalance in the data, we used Synthetic Minority Over-sampling Technique for Nominal and Continuous (SMOTE-NC) on the training data and random under-sampling on the test data. The classifiers were hyperparametrized with Bayesian optimization to avoid exploring the whole hyperparameter space. Ordinal logistic regression had slightly better A U C and slightly worst accuracy than LightGBM. Nevertheless, a boosting algorithm applied to the ordinal logistic regression classifier slightly improved its performance over LightGBM, both in AUC and accuracy.

The incorporation of ICT in university teaching has been constantly growing. Hundreds of tools, platforms, apps, and other technological products are made available to students and teachers in order to improve the training process. However, there is a great weakness in teachers in relation to their pedagogical training, so the use of ICTs does not guarantee a quality training process. This article presents a collaborative platform that facilitates the design of learning modules. This platform offers a pedagogical scaffolding that tries to overcome some deficiencies in the area of instructional design. This scaffolding incorporates three elements in the conception of the design of a module: i) evaluation as the guiding axis for the design, ii) Bloom's taxonomy to guide the definition of the module's learning outcomes, and iii) the conversational framework for facilitate the design of the different tasks that the student must perform. The platform was tested by a group of teachers from different disciplines. At the end of the experience, a focus group was carried out on the participants, which yielded favorable results, which are summarized in that teachers with less experience benefit from the experience of other teachers, and the collaborative work that the same platform offers is also enhanced.

The practice of teaching to train the professionals of the future is increasingly demanding. Many of the trainers in Higher Education Institutions do not have a training in pedagogy, so they do their best to achieve good results in their students. Teachers base their methodology on repeating patterns learned, training received, formal and informal support from pedagogical support units in universities, among other elements. This article presents a platform that supports the design of pedagogical strategies, which is based on three pillars. First, planning based on pedagogical criteria (Bloom's taxonomy and conversational framework). Second, a design that arises from assessment, understood as an always formative activity. And third, collaboration both in the design of the activities and in sharing these designs with others. This platform has been used for designs in the computing area with positive evaluations from the participating teachers. Teachers from other disciplines have also joined, confirming the usefulness and transversality of the platform. A direct impact on students with more solid and relevant learning is expected.

Teaching in higher education institutions is being developed with extensive use of technology. One of its consequences is the transformation of traditional teaching practices, allowing technological resources to be incorporated into the teaching-learning process that are made available to students to promote specific learning and potentially allow them to develop transversal skills. Higher education institutions have made progress in developing educational models of competency training. Until now, the greatest difficulty lies in implementing learning activities that allow both developing disciplinary learning and modeling transversal skills (generic competences) for professional life. This work reports an experience based on the incorporation of virtual learning objects, developed under the SCORM standard arranged in an LMS Moodle, in order to develop meaningful learning and to model generic competences in a group of students belonging to the Catholic University of Temuco.

In this paper, we present a new methodology for the assessment of Programming and Data Structures courses in conditions of Covid-19. The study is carried out at the Catholic University of Temuco, based on the suggestions received at the beginning of the pandemic, by students of the Computer Engineering career. The proposal consists of redirecting the assessment to freely accessible online programming platforms. As part of the study, a mapping of the contents to be evaluated in both subjects with problems available on platforms such as LeetCode and CodinGame is carried out. We manage to cover all the content to be evaluated using these platforms. An analysis of the results of the course, applying this new methodology, is made. Through a survey, applied at the end of the semester, we collected information about the perception of students, concerning aspects related to the changes in the assessment of the course. The results of the survey are analyzed qualitatively and quantitatively, perceiving improvements in the promotion indicators and the motivation of the students.

Currently, efforts to improve teaching and learning processes in higher education are increasingly being seen. New student-centered approaches, technological support, the heterogeneity of students, are some elements that are being considered in new strategies. On the other hand, there is a large mass of teachers without pedagogical training who tend to replicate models acquired in their own training, which are quite far from current trends. This generates inappropriate pedagogical designs, focused on a traditional training. This article presents a learning module design approach that relies on assessment as its foundation. This approach includes a collaborative platform that facilitates design, especially for non-pedagogical teachers. The platform was validated with five teachers, obtaining satisfactory preliminary results.

Automatic recognition of visual objects using a deep learning approach has been successfully applied to multiple areas. However, deep learning techniques require a large amount of labeled data, which is usually expensive to obtain. An alternative is to use semi-supervised models, such as co-training, where multiple complementary views are combined using a small amount of labeled data. A simple way to associate views to visual objects is through the application of a degree of rotation or a type of filter. In this work, we propose a co-training model for visual object recognition using deep neural networks by adding layers of self-supervised neural networks as intermediate inputs to the views, where the views are diversified through the cross-entropy regularization of their outputs. Since the model merges the concepts of co-training and self-supervised learning by considering the differentiation of outputs, we called it Differential Self-Supervised Co-Training (DSSCo-Training). This paper presents some experiments using the DSSCo-Training model to well-known image datasets such as MNIST, CIFAR-100, and SVHN. The results indicate that the proposed model is competitive with the state-of-art models and shows an average relative improvement of 5% in accuracy for several datasets, despite its greater simplicity with respect to more recent approaches.

The kinesiological treatment of hand typically requires exercise either of mobility, elasticity or reflexes in the hands. This exercises requires to be evaluated by an expert, however, he could not be present. An alternative is the use of technologies, where eventually these exercises could be remotely modelled using 3D models for a more detailed representation of action respect to typical video records. Nonetheless, the hands are a very complex object as it is conformed by multiple parts, which difficults the application of typical sensors as RGB cameras. In this work, we propose to develop a software capable of implementing hand tracking system over a real-time infrared device. This system associates the sensor measures to precise position and pose of the hand. Besides, the software graph the hands three-dimensionally in a virtual environment considering different exercises that the user might perform. We expect that this proposal will be the basis of an application for kinesiological exercise in the hands, since this implementation technology also aims to increase the motivation of users to carry out the exercises and consequently, this could even mean a decrease in the recovery time.

In this work, we implemented a CUDA parallelized simulated annealing algorithm to solve the student-school assignment problem in a highly segregated environment. The objective function optimized considered the average distance from the students to their assigned school, the socio-economic segregation via the dissimilarity index, and the cost of schools partially filled. Using data from the MINEDUC, the INE, and the Municipality of Temuco (Chile), we simulated the distribution of Temuco's student population, solving its students' assignment to the city's schools (29853 students to 85 schools). The results obtained were better with a high number of block (simultaneous students exploring), and a low number of threads (simultaneous schools explored by these students) instantiated in the GPU. Algorithm execution time worsens with the number of blocks and the number of threads, although it remained below 1000 seconds in the worst and below 400 seconds in the best case. However, the algorithm achieves excellent results in reducing socio-economic segregation, taking it from a high level to almost making it disappear. We achieved this result, even with a reduction of the average distance from students to their assigned school.

The teaching and learning process is still complex in the engineering area. Traditionally there are imperative courses with a high failure rate because of their complicated topics. This work presents a model of learning support with two fundamental aspects: technological support and peer learning. First, the technological platform is a multi-tooled social network. On the other hand, peer learning includes collaboration, bonding, feedback, and autonomous learning, as fundamental aspects of its implementation. Another crucial element of this model is the active participation of the assistants, who are the peers trained to guide the students in the process. This mentoring takes place in two environments, face-to-face and virtual. The face-to-face environment gathers all the infrastructure and human resources of the Colegio de Ayudantes (institutional unit), while the virtual environment contemplates the social network along with a set of tools that support this model. There are initial results of validation of the platform regarding its use and potential. Although this implementation starts in imperative courses of the faculty, it can be easily expanded to other courses of the university. This implementation can be possible due to the unique characteristics of a social network that self-regulates its growth.

The pandemic context forced most of the Higher Education Institutions, and therefore their faculties, to face a semester of teaching in an online mode. The universities defined general guidelines and strategies to try to guarantee a minimum quality of the process. However, the literature reports that, depending on its particularities, preparing a semester-long online course requires a design of several months. Therefore, each lecturer worked according to her experience, guidelines, and support she received: "What could be done, was done." This article reports the teaching experience of the 1st semester 2020 of the Computer Science undergraduate program at the Universidad Católica de Temuco. Some lecturers of this undergraduate program had previous experience in online training; hence they designed strategies that differed in teaching, evaluation, and interaction methodologies while using several technologies available to them. A sample of 115 of our students answered a questionnaire to assess the preferences and problems during this semester. The results show a heterogeneous set of clusters. However, they point to a constructivist asynchronous vision in a group of our students that prefer short videos uploaded by their professors, and the resources that, by their means, they find on the Internet, point towards a constructivist asynchronous vision in a group of our students. The results also allowed us to rescue good practices that help in proposing an emergency online teaching model that will guide the design of the second-semester courses.

la experiencia de la Facultad de Ciencias Económicas y Administrativas de la Universidad Austral de Chile en implementar un programa que fomenta el emprendimiento y el desarrollo de habilidades para emprender entre sus estudiantes de pregrado.Como metodología, el programa ha utilizado métodos de reciente uso en el contexto universitario a través de la creación de una asignatura que facilita el desarrollo de proyectos de emprendimiento con un enfoque social, colaborativo e interdisciplinario.De los resultados de la experiencia, se menciona la generación de 39 proyectos estudiantiles en el período de análisis entre otras iniciativas realizadas por el programa.Se concluye que la experiencia puede aportar a la discusión y diseño de este tipo de experiencias en las universidades chilenas.

La tradición “bancaria” de la enseñanza de derecho y las formas tradicionales de evaluación basadas en la exposición fiel de contenidos, han sido ampliamente criticadas en la literatura. Como estrategia disruptiva de esta tradición, en este trabajo se expone la aplicación de la Evaluación Basada en la Certeza (CBM) en un curso obligatorio de primer año de la carrera de derecho de una universidad del sur de Chile, durante el año académico 2018-2019. Mediante la CBM se busca pasar del mero “recuerdo” de conocimientos a la “evaluación” de los mismos. Para lograr este objetivo, las/los estudiantes, mediante un modelo de aprendizaje autorregulado, deben autoevaluarse tanto con retroalimentación interna como externa. Los resultados principales alcanzados muestran que los alumnos aplican procesos de autorregulación y autoevaluación al seleccionar el nivel de certeza que maximiza su puntaje esperado. Mediante un modelo mixto con efectos aleatorios, controlando por efectos fijos de prueba, podemos confirmar que la probabilidad de responder correctamente una pregunta tiene como predictores significativos la retroalimentación, la longitud de la prueba y el nivel de certeza de la respuesta. Por contra, el género del/de la estudiante no es un predictor significativo, ni aisladamente, ni como término de interacción con otras covariables.

Purpose The purpose of this paper is to evaluate Chinese indirect trade relations in the global trade network to observe if the objectives identified by Cai (2017) in the Belt and Road Initiative (BRI) are being fulfilled, especially with Latin America and the Caribbean (LAC) countries. Design/methodology/approach Using data from the UNCTAD (2016) for the period 2011–2015, a normalized exports network is built. It is analyzed with the Forgotten Effects Theory and the PageRank algorithm. A Monte Carlo experiment with 10,000 replicates is performed to account for its volatility. Findings The paper identifies one instance in which China's peripheral countries are importing raw materials and commodities -–oil products – to produce low technological value-added products, which, in turn, are exported to China. LAC countries do not have significant indirect trade relations with China when the former is the origin country, while the latter is the destination in a trade relationship. The trade network has a clear core-periphery structure, with China belonging to its core, although being only the fourth most central node in the network. Originality/value This paper contributes with both a new methodology for the analysis of indirect trade relations and the results found for China under the BRI and its trade relationship with LAC economies.

Abstract In recent years, the pollution problem has gained great importance due to its socioeconomic implications for people regarding health or logistic issues. The pollution level classically is measured with specialized expensive detectors located in some few locations. In the case of Temuco city there are three such centralized pollution monitoring stations. An alternative approach for measuring the pollution level of cities makes use of inexpensive pollution sensors located on public transportation vehicles. Nonetheless, a drawback of this approach is that these inexpensive sensors can be sensitive to noise, vehicle movement, human intervention or technical failures. Therefore, it is relevant to be able to automatically detect inaccurate or failing sensors as they are multiple and cannot be submitted frequently to a technical revision. In this work, we propose a method to automatically detect these anomalous sensors by an unsupervised deep learning approach using autoencoders. This work is part of an ongoing project where massive data are not still available. In this context, the simulated output of mobile pollution sensors is generated by a time series model that systematically inserts outlier measurements. Our results indicate that the proposed detection method is able to reliably reproduce the data generated and to detect the simulated outliers with an accuracy of over 95%. A post-publication change was made to this article on 3 Jul 2020 to correct an author name.

The aim of this study is to examine the incidence of economic activity on soft innovation in the mining sector. Through a global analysis of indirect incidents and using the theory of forgotten effects. Thus, the case of the mining industry in Chile was analyzed, given that it is one of the industries that contributes the most to GDP in Chile. The empirical study was carried out through the application of a structured survey towards three experts from the mining sector. In addition, the study concludes with evidence that oil price has a direct incidence on investments in mining and economic expectations, and indirectly in average income middle managers, market share of the company in the mining sector, and growth of imports and exports in the mining sector.

S3E2 is a web-based geographic information system (GIS) designed for the visualization and analysis of the socioeconomic segregation of Chile’s elementary education system of Chile. It consists of a frontend developed in JavaScript using ReactJS, React-Redux, Leaflet and D3.js, an API developed in Go and ECHO, and a documentary database man- aged with MongoDB. Data comes from Chile’s Ministry of Education, while the provisions of Law 20.248 serve as indicators of vulnerability. S3E2 graphically shows different segregation indices found in the literature, calculated at the commune level. It also allows visualizing, at this same level, the educational institutions that compose it, their basic information, and time series associated with them. S3E2 is a flexible and fast web-based GIS, with a low cost of implementation, due to the usage of free software -or at least free licensing- tools, thus serving as a template for new web-based GIS in different contexts.

Currently, a large amount of data is generated in the telemetry sector of vehicles in cities due to the continuous monitoring of vehicle trajectories through multiple sensors. Some trajectories generated by the sensors turn out not to correspond to the reality due to artefacts such as buildings, bridges or sensor failures, and where due to their large volume a manual verification of their correctness is not feasible. In this work, we propose the use of deep neural network models without supervision based on stacked autoencoders to detect atypical trajectories in vehicles within Santiago, Chile. The results show that the proposed model shows that it is able to detect that the atypical vehicle paths detected are at least 85% correct when considering the validation of a human expert. As future work, we propose to incorporate the use of LSTM networks in our model.

The reduction of student segregation should be one of the priorities of any state in terms of educational legislation and regulation since there is empirical evidence that students belonging to families in vulnerable situations, who attend less segregated schools, tend to have better academic results. In this work, we implement the reaction-diffusion model to simulate decentralized student enrollment. Reaction-diffusion equations are used in conjunction with the finite volumes method to generate a discrete domain of space over which students search for a suitable school and the explicit Euler method for the approximation of time. The computational implementation, through CUDA, allows solving for the problem of wide territories, such as cities, taking advantage of the performance benefits offered by parallel computing by GPU at a low cost. In our tests, simulations have a low runtime, converging to the total assignment of students to schools.

Peer learning (PL) is a strategy where students learn from each other without direct teacher intervention. Five pillars support this strategy: interaction, negotiation, organization, feedback, and evaluation. According to the problems students face, self-learning is one of the tools they need to face and is a generic competence they should apply. This article presents a PL experience in the undergraduate degree in Computer Science Engineering (CSE) of the Catholic University of Temuco (UCT), emphasizing how Information and Communication Technologies (ICT) can support each pillar. The curriculum includes a course called Integration Workshop (IW), a capstone course to instantiate the competences acquired by the students up to that point. In this experience, we joined teams from IW II, as developers, and IW IV, as SCRUM Masters. Students rated the experience positively while highlighting the development of other generic competences such as collaborative work, knowledge management, and valuation and respect for diversity.

Teaching computer programming is a complex task. In computer science programs students usually have big difficulties dealing with programming competence. In regional universities in Chile, this problem tends to increase, because the gaps students have before enroll computer science programs. In order to address this problem, a course in the first year of the curriculum in Computer Science in Temuco Catholic University, has been incorporated. This course supports teaching of computer programming using Lego NXT robots and the NXC language. The methodology additionally supports the generic skills of creativity and innovation, teamwork and written communication. The results obtained are satisfactory, since students can experiment into real and recreational elements, programming bases, obtaining interesting and motivating products early. In addition, students achieve an early approach to advanced concepts such as concurrent programming, threads, multitasking, among others.

This paper proposes a synthetic indicator that measures the degree of development of Chilean universities, by quantitatively assessing each institution's missions (research and teaching). A model of structural equations solved through partial least squares is used, analyzing data from 54 universities in Chile, from the year, 2017. The results show that there are important differences among universities regarding the degree of development they present, according to the measure of the intensity with which they perform research and teaching activities. The article proposes public policy measures linked to improving the university education system that would help reduce the gap between the lowest performing universities and those that lead the system.

Today, there is growing interest in the automatic classification of a variety of tasks, such as weather forecasting, product recommendations, intrusion detection, and people recognition. "Mixture-of-experts" is a well-known classification technique; it is a probabilistic model consisting of local expert classifiers weighted by a gate network that is typically based on softmax functions, combined with learnable complex patterns in data. In this scheme, one data point is influenced by only one expert; as a result, the training process can be misguided in real datasets for which complex data need to be explained by multiple experts. In this work, we propose a variant of the regular mixture-of-experts model. In the proposed model, the cost classification is penalized by the Shannon entropy of the gating network in order to avoid a "winner-takes-all" output for the gating network. Experiments show the advantage of our approach using several real datasets, with improvements in mean accuracy of 3-6% in some datasets. In future work, we plan to embed feature selection into this model.

Within the increasing automation of tasks, it is necessary to obtain relevant information from images, to have clarity of the actions that must be performed. In this process, it is possible to detect objects individually, however this mode can generate a considerable error due to the enormous number of ways in which an object can be presented. It is therefore necessary to improve the level of accuracy, and a way to achieve this is taking into account the relationships between objects as well. Until now, context-oriented models generate relationships between objects without discriminating between input images. It is necessary that the model operates on each particular image, to reduce the error and obtain a conclusive result. In response to the above, this work proposes an adaptive probabilistic context-model. The model in question is a functional that processes the occurrence probability of the objects in each image, and generates the relations between the detectable objects using a Bayesian network in the form of a tree. This model is updated with the features of each image, requiring a minimization of the quadratic error through a numerical approximation, obtained by Newton-Raphson method. Comparisons were made between different heuristic proposals, as well as tests on different context-oriented databases, in order to validate the results. On the other hand, tests were performed using features extracted through histograms of orient gradients and Deep Learning. It was found that an improvement in the prediction of the order of 20% is feasible in the testing process.

Self-regulation on the part of students regarding the use or adoption of technologies is generally directed as a process of dynamic transitions between roles in life, the environment, globalization and ICTs. The objective of this study was to provide a framework to determine the interrelation of activation of the use of ICTs by students in the context of the evaluation of general competences through project-oriented learning. We worked with a sample of 93 students. An experimental design of random sampling was used. In the analysis of results, the degree of use of ICTs according to the convergence model is demonstrated [1]. Consequently, self-regulation training facilitates students to use strategies related to reflective learning in their own personal work environment and collaborative network work environment.

Global changes and current trends have led to major revisions in the field of education. In the context of higher education, it has proliferated successfully to move the curricula to a competency-based education. Also, the incorporation of ICT in the training process is a growing trend that has been strongly positioned in all levels of education. However, beyond the models and paradigms, what is relevant is the implementation of an innovative proposal. This is where the problems of a complex implementation are experienced. At UC Temuco, the computer science program renewed its curriculum to a competency model. For the evaluation of achievement of competences, a course called integration workshop is included in the curriculum at the end of each year. These courses include a series of individual and group strategies, which have been designed, refined and improved over the years. ICT is a strong component in the support of strategies, including tools for collaboration, authorship, software project management, among others. The workshops are evaluated externally through an event called "project fair", where actors from the business, government and academic world are invited. Results have been very satisfactory and have allowed a continuous improvement of the strategies and the selection of ICT support tools.

En este trabajo nos proponemos dos objetivos: primero, exponer cómo la novedad, como requisito para la obtención de una patente, obstaculiza la realización de una ciencia abierta en términos de comunicación y uso; segundo, determinar los incentivos que llevan a la industria a optar por una ciencia abierta en los términos propuestos, dejando de lado el patentamiento. Para lograrlos, primero presentaremos los argumentos jurídicoeconómicos que llevan a considerar la patente como incentivo del desarrollo tecnocientífico, pero cómo el “efecto red” es coartado por la misma. Segundo, analizaremos el período de gracia como excepción que impide que la divulgación de la invención, previa a la solicitud de una patente, no destruya la novedad. Finalmente, estudiaremos cómo la industria acoge y alienta al software libre, caso paradigmático de la ciencia abierta, al identificar que el “efecto red” genera una utilidad superior a la de otras estrategias de producción científica.

Este trabajo tiene como objetivo analizar el comportamiento de la ocupacion en el mercado del trabajo en Chile a nivel regional, determinando las diferencias y similitudes entre regiones, y de las regiones con el agregado de empleo a nivel nacional. Para la realizacion de dicho trabajo se han utilizado las tasas mensuales de ocupacion regionales y a nivel pais para el periodo 1986-2010, y utilizado metodos de Clusterizacion de series temporales para identificar conjuntos de regiones con comportamientos similares. Los resultados muestran la existencia de diferentes regimenes en los mercados laborales de las regiones y el relevante rol de algunas regiones en el comportamiento del agregado nacional. The objective of this work is to analyze the employment behavior in the labor market in Chile at the regional level, determining the differences and similarities between regions, and of the regions with the aggregate employment at the national level. For this work, we studied the monthly regional employment rates during the 1986-2010 period for Chile by applying time series clustering methods to identify regions with similar dynamics. Afterward, we use that information to build a model that explains the contributions from the regional markets to the national aggregate. Results show the existence of regimes on the local labor markets and the relevance of the Metropolitan Region on the country-level behavior.

RESUMEN: El objetivo de este trabajo ha sido analizar el grado de sincronismo entre el ciclo económico de la Región de Magallanes respecto del ciclo económico nacional para el período 1985-2010. Para determinar el ciclo económico se ha usado el procedimiento de Harding y Pagan (2002) y medido el sincronismo con el índice de contingencia de Pearson. Nuestros resultados muestran evidencia de diferentes grados de sincronismo entre los ciclos económicos regionales respecto del ciclo económico nacional, destacando la región de Magallanes por su menor y mayor volatilidad de su crecimiento económico y por presentar un bajo grado de sincronismo entre su ciclo económico y el ciclo económico nacional, como también con los ciclos de las demás regiones.

The breakthrough of information and communication technologies and their smooth incorporation into daily living has brought a cultural change in both, society and education. The traditional and formal paradigms and contexts represent new challenges of adjustments that consider the new kind of student that goes to university classrooms. Thus, the educational model from Universidad Católica de Temuco considers an intensive use of ICT in the educational process, and a gradual growth of the autonomy of students during training. In the Computer Science program, these elements of the model have been partially addressed through the incorporation of e-learning modules that foster autonomy. A technological platform was implemented to support an e-learning methodology. The methodology used in the study is described in this article. The outcomes obtained in courses where the methodology was applied were successful, which made possible new initiatives that include the suggested improvements from the assessment of the conducted experiences.

Today, organizations are beginning to realize the importance of using as much data as possible for decision-making in their strategy. The finding of relevant patterns in enormous amount of data requires automatic machine learning algorithms, among them, a popular option is the mixture-of-experts that allows to model data using a set of local experts. The problem of using typical learning algorithms over Big Data is the handling of these large datasets in primary memory. In this paper, we propose a methodology to learn a mixture-of-experts in a distributed way using PETUUM platform. Particularly, we propose to learn the parameters of mixture-of-experts by adapting the standard stochastic gradient descent in a distributed way. This methodology is applied to people detection with standard real datasets considering accuracy and precision metrics among other. The results show a consistent performance of mixture-of-experts models where the best number of experts varies according to the particular dataset. We also evidence the advantages of the distributed approach by showing the almost linear decreasing of average training time according to the number of processors. In a future work, we expect to apply this methodology to mixture-of-experts with embedded variable selection.

In these days, there are a growing interest in pattern recognition for tasks as prediction of weather events, recommendation of the best route, intrusion detection or face detection. Each of these tasks can be modelled as classification problem, where a common alternative is to use an ensemble model of classification. A well-known example is given by Mixture-of-Experts model, which represents a probabilistic artificial neural network consisting of local experts classifiers weighted by a gate network, and whose combination creates an environment of competition among experts seeking to obtain patterns of the data source. We observe that this architecture assume that one gate influence only one data point, consequently the training can be misguided in real datasets where the data is better explained by multiple experts. In this work, we present a variant of regular Mixture-of-Experts model, which consists of maximizing of the entropy of gate network in addition to classification cost minimization. The results show the advantage of our approach in multiple datasets in terms of accuracy metric. As a future work, we plan to apply this idea to the Mixture-of-Experts with embedded feature selection.

In the digital era, tech devices (hardware and software) are increasingly within hand's reach. Yet, implementing information and communication technologies for educational contexts that have robust and long-lasting effects on student learning outcomes is still a challenge. We propose that any such system must a) be theoretically motivated and designed to tackle specific cognitive skills (e.g., inference making) supporting a given cognitive task (e.g., reading comprehension) and b) must be able to identify and adapt to the user's profile. In the present study, we implemented a feedback-based adaptive system called A-book (assisted-reading book) and tested it in a sample of 4th, 5th, and 6th graders. To assess our hypotheses, we contrasted three experimental assisted-reading conditions; one that supported meta-cognitive skills and adapted to the user profile (adaptive condition), one that supported meta-cognitive skills but did not adapt to the user profile (training condition) and a control condition. The results provide initial support for our proposal; participants in the adaptive condition improved their accuracy scores on inference making questions over time, outperforming both the training and control groups. There was no evidence, however, of significant improvements on other tested meta-cognitive skills (i.e., text structure knowledge, comprehension monitoring). We discussed the practical implications of using the A-book for the enhancement of meta-cognitive skills in school contexts, as well as its current limitations and future developments that could improve the system.

Pedestrian detection has multiple applications as video surveillance, automatic driver-assistance systems in vehicles or visual control of access. This task is challenging due to presence of factors such as poor lighting, occlusion or uncertainty in the environment. Deep learning has reached many state-of-art results in visual recognition, where one popular and simple variant is stacked autoencoders. Nonetheless, it is not clear what is the effect of each stacked autoencoders parameter in pedestrian detection performance. In this work, we propose to revise the feature representation for pedestrian detection considering the use of deep learning using stacked autoencoders with a sensitivity analysis of relevant parameters. Additionally, this paper presents a methodology for feature extraction using stacked autoencoders. The experiments show that this model is capable of creating a meaningful visual descriptor for pedestrian detection, which improves the detection performance in comparison to baseline techniques without an optimal setting of parameters. In presence of occlusion or poor people images, we found diffuse and distorted visual patterns. A future avenue is the learning of the degree of noise for improving the generalization capabilities of the learned features.

The high rate of university dropout and low graduation rates are very relevant social problems today. Since there are many possible causes of desertion and university graduation, in this paper, we propose to find, analyze and weigh the factors that allow predicting if a student will drop out or graduate according to prior information available using data mining techniques and statistical models. We will focus in the case of Catholic University of Temuco, using real data from that institution. This study reveals relevant variables in opinion of human experts, which demonstrates the ability of automatic models to represent the dropout and graduation at the university.

The development of generic skills in engineering education is a demand already established by the industry. In the Unversidad Católica de Temuco, since 2008, a new educational model has been defined, which is a competency-based model. Therefore, each curriculum must define their programs in terms of both specific and generic skills. One of the biggest problems presented by the implementation of the model has been the process of validation of generic skills. This article reports a strategy that facilitates the faculty the validation process of generic skills. This strategy considers two key elements; the use of ICT and the generation of a virtual community of practice. In this community, faculty can access activities uploaded by their peers and can draw on the experience of these, as well as contribute with their own experience.

Resumen en este trabajo se informan los resultados de una investigación cuyo objetivo fue comprobar la eficacia del programa estatal It's my Turn, implementado por el ministerio de educación de Chile, para el autoaprendizaje del inglés en contextos rurales de todo el país.se midió el aprendizaje de 120 estudiantes de 10 escuelas rurales de la Región metropolitana (Rm) con pruebas de entrada y salida.Asimismo, se realizaron observaciones para examinar la utilización del recurso, y un cuestionario a los docentes para conocer su percepción del programa.una prueba t arrojó una diferencia significativa entre las pruebas de entrada y salida (p < 0,0001), lo que respalda la efectividad del programa.Las observaciones pusieron en evidencia métodos heterogéneos de aplicación y los cuestionarios mostraron que los profesores tienen diferentes perspectivas acerca de la eficacia de It's my Turn.La combinación de resultados cuantitativos y cualitativos apunta a la efectividad del programa, pero también a la necesidad de realizar cambios en su implementación.

The neutral nature of Internet has allowed its consolidation as a crucial tool in the dissemination of knowledge and access to culture. Due the creation of new business models of Internet access, a debate about the need of keeping a neutral Internet has emerged, leading to a sudden regulatory process that seems to emerge from a process of public consensus. However, participation in this debate requires knowledge in telecommunications, economics, and law, leaving participation to expert actors. In public consultations on Net Neutrality and in the resulting legal documents, three fundamental problems related to net neutrality are studied. Firstly, what constitutes a neutral, open and free Internet? Secondly, what is the effect of traffic management and what are the consequences of providing differentiated services? Finally, can transparency be an efficient tool to curb potential violations of net neutrality? This article presents the historical background that led to this debate and how its main points have been treated primarily in USA and Europe.

El Yali National Reserve is a protected area that is priority number one for the National Forestry Corporation (CONAF). It is located in the V region of Chile, where the risk of wildfire is quite high. In this context, training actors involved in combating wildfire is a relevant issue. Under the simulation of serious video games, a simulator that allows this training was built, since it is not possible to recreate realistic scenarios for this kind of training. Current problems in this context are decision-making with incomplete information, and the non-lineal consequences of these decisions. In the first stage of the project the terrain model was built and subjected to a validation process by experts at the University of Playa Ancha. Preliminary results are satisfactory and allow progression to the next stages of the project.

The growing application of simulation tools to increasingly complex problems makes the use of validation techniques essential to improve confidence in the veracity of those simulation results. The feature selective validation (FSV) method is widely used today, because of its versatility and simplicity. However, despite its many advantages, some problems have been identified in the standardized FSV method. Those drawbacks can produce misleading results for the validation process and constitutes enhancement opportunities for further research on FSV improvement. This paper presents a review that summarizes the major problems of the FSV and their possible solutions as a justification to reform the FSV method and also looking towards the forthcoming update of the standards IEEE 1957.1/2.

International audience

The research objective of this paper is to find out the relation between the influence of a demand point on a center (a facility) location and the distance from this point to this center, according to a given metric, that is to say in our case a Lp-norm. Firstly exposed in a theoretical isotropic space, the relation is discussed and applied to a simple virtual transport network, using intermediate values of p, such as p=1.5 or p=3. Since the individual influence can be mapped, this approach provides to planners an interesting insight into on the impact of the metric on the relation between the center and the demand point influence spatial distribution.

In this paper, we present a tool to help reduce the uncertainty presented in the resource selection problem when information is subjective in nature. The candidates and the ?ideal? resource required by evaluators are modeled by fuzzy subsets whose elements are trapezoidal fuzzy numbers (TrFN). By modeling with TrFN the subjective variables used to determine the best among a set of resources, one should take into account in the decision-making process, not only their expected value, but also the uncertainty that they reflect. Respecting this condition, for each candidate an asymmetric index evaluates the distance between the TrFNs for each of the variables and the corresponding TrFNs of the ?ideal? candidate, consolidating them through a weighted average that lets the decision-maker make the final comparison between the candidates, and the selection of the one best suited. We apply this contribution to the case of the selection of the product that is best suited for a ?pilot test? to be carried out in some market segment.

La tendencia natural de la sociedad a satisfacer sus necesidades cotidianas, la ha llevado a adoptar nuevas formas de interacción apoyadas en tecnologías de computación ubicua, que brindan una nueva dimensión de conectividad. Internet de Objetos (Internet of Things [IOT]) es uno de los nuevos conceptos ligados a estas tecnologías. Junto con el surge el de Universidad de Objetos, como una alternativa que permite la interacción entre personas y objetos comunes en escenarios educativos. Bajo este concepto –y con base en los criterios del Modelo Conceptual de Internet de Objetos de Aprendizaje–, se diseñaron e implementaron actividades e-Campus, en varios escenarios, con el propósito de evaluar la aplicación o la pertinencia de la tecnología Near Field Communication en el ambiente universitario. Un riguroso análisis estadístico de los resultados permite concluir que las actividades propician un espacio de interacción satisfactorio para los participantes, mejorando la percepción de usabilidad de los ambientes inteligentes pertenecientes a la IoT.

In this paper we revisit the Textual Energy model. We deal with the two major disadvantages of the Textual Energy: the asymmetry of the distribution and the unbounded ness of the maximum value. Although this model has been successfully used in several NLP tasks like summarization, clustering and sentence compression, no correction of these problems has been proposed until now. Concerning the maximum value, we analyze the computation of Textual Energy matrix and we conclude that energy values are dominated by the lexical richness in quadratic growth of the vocabulary size. Using the Box-Cox transformation, we show empirical evidence that a log transformation could correct both problems.

"La tendencia natural de la sociedad a satisfacer sus necesidades cotidianas, la ha llevado a adoptar nuevas formas de interacción apoyadas en tecnologías de computación ubicua, que brindan una nueva dimensión de conectividad. Internet de Objetos ( Internet of Things [IOT]) es uno de los nuevos conceptos ligados a estas tecnologías. Junto con el surge el de Universidad de Objetos, como una alternativa que permite la interacción entre personas y objetos comunes en escenarios educativos. Bajo este concepto –y con base en los criterios del Modelo Conceptual de Internet de Objetos de Aprendizaje–, se diseñaron e implementaron actividades e-Campus, en varios escenarios, con el propósito de evaluar la aplicación o la pertinencia de la tecnología Near Field Communication en el ambiente universitario. Un riguroso análisis estadístico de los resultados permite concluir que las actividades propician un espacio de interacción satisfactorio para los participantes, mejorando la percepción de usabilidad de los ambientes inteligentes pertenecientes a la IoT."

In this paper, the authors present a methodology to solve the weighted barycenter problem when the data is inherently fuzzy. This method, from data clustered by expert visual inspection of maps, calculates bi-dimensional fuzzy numbers from the spatial clusters, which in turn are used to obtain the weighted fuzzy barycenter of a particular area. The authors apply the methodology, to a particularly apt data set of forest fire breakouts in the PACA region of southeastern France, gathered from 1986 to 2008, and sliced into five periods over which the fuzzy weighted barycenter for each one is obtained. Two weighting schemes based on fire intensity and fire density in a cluster were used. The center provided with this fuzzy method provides leeway to planners, which can see how the membership function of the fuzzy solution can be used as a measurement of “appropriateness” of the final location.

The increasing application of simulation tools to increasingly complex problems makes the use of validation tools essential to improve confidence in the veracity of those simulation results. IEEE Standard 1597.1 is the first true standard for the validation of computational electromagnetics method. This standard uses the feature selective validation (FSV) method as the key quantification tool. However, despite its many advantages, there have been some interesting issues surrounding the validation of transients. This paper presents a new approach to the validation of a set of generally representative transient types using the FSV method and shows how the previously experienced limitations can be overcome. In order to analyze the main parameters associated with transient comparison, a survey which included 20 experts was conducted. This information was used to identify the significant regions that need to be taken into account in the transient comparison. Finally, using the statistics obtained by the experts, a new solution was defined and its improvement over the existing approach was demonstrated.

The competences-based training arises from the company as an educational strategy that addresses the shortcoming claimed by the labor market, in relation to the poor development of many skills in computer science engineers. Some strategies that allow to validate soft skills in a competency-based educative model for a computer science curriculum are described. The preliminary results report retention rates of 77.66% in the first year and 81.66% in the second year, and the pass rates of the first year was 56.12% and 69.10% in the second year.

Effective communication among first responders during response to natural and human-made large-scale catastrophes has increased tremendously during the last decade. However, most efforts to achieve a higher degree of effectiveness in communication lack synergy between the environment and the technology involved to support first responders operations. This article presents a natural and intuitive interface to support Stigmergy; or communication through the environment, based on intuitively marking and retrieving information from the environment with a pointer. A prototype of the system was built and tested in the field, however the pointing activity revealed challenges regarding accuracy due to limitations of the sensors used. The results obtained from these field tests were the basis for this research effort and will have the potential to enable communication through the environment for first responders operating in highly dynamical and inhospitable disaster relief environments.

In this paper, we present a methodology to solve location problems when the data used is inherently fuzzy. This method, from dataclusterized with the fuzzy c��means algorithm, calculates bi-dimensional fuzzy numbers from the clusters which are used to calculatea fuzzy solution. We apply the methodology, with different objective functions, to a particularly apt data set of forest fire breakouts inthe Bouches du Rhone region of southern France, gathered from 1981 to 2009. The robustness of the method is then evaluated witha Monte Carlo simulation in which the number of clusters change. The solution provided with this fuzzy method provides leeway toplanners, which can see how the membership function of the fuzzy solution can be used as a measurement of appropriateness of thefinal location.

International audience

Remote experimentation laboratories (REL) are systems based on real equipment that allow students to carry out a laboratory practice through the Internet on the computer. In engineering, there have been numerous initiatives to implement REL over recent years, given the fundamental role of laboratory activities. However, in the past efforts have concentrated on laboratory groups interacting face to face, disregarding the capacities of distributed student collaborative environments. This article proposes a model for the implementation of REL in a distributed collaborative scenario, focusing on two crucial key elements: shared knowledge and interaction for collaboration. The model focuses on the methodological aspects of executing REL in a distributed collaborative scenario and disregards technical aspects of the implementation. This study analyses distributed collaborative scenarios where the teacher plays a fundamental role in REL configuration to ensure group collaboration. The new model introduced presents diverse aspects that are associated with the methodological implementation of REL in the field of engineering; hence it is to be regarded as a foundation for teachers developing REL in distributed collaborative scenarios.

Our goal is to achieve deeper understanding of the limitation of queueing models, on one hand, and of common simulation practice, on the other hand, as tools for predicting performance of bandwidth sharing between competing TCP flows. In particular, we (i) present an overview of simulation pro

The initial aim with the MeAdian was to obtain a local spatial filtering method which enables to process a robust centre based on a combination of the mean and the median. The MeAdian is an auto-adaptive filter that tends to the mean when this one is more robust, to the Median otherwise. The MeAdian, including or not the covariance of the Mean and the Median, remains one of the most robust estimators faced to different distributions, due to its auto-adaptive capabilities. In this paper, we improve the MeAdian filtering for contour detection in image analysis. The results show the double effect of the MeAdian: a combination of smoothing and planing according to the local distributions encountered. The MeAdian tends to define areas with high homogeneity, like the median does, but whose borders are smoothed or antialiased, like the mean does.

Internet growth has allowed unprecedented widespread access to cultural creation including music and films, to knowledge, and to a wide range of consumer information. At the same time, it has become a huge source of business opportunities. Along with great benefits that this access to the Internet provides, the open and free access to the Internet has encountered large opposition based on political, economical and ethical reasons. An ongoing battle over the control on Internet access has been escalating on all these fronts. In this paper we describe first some of the ideological roots of free access to the Internet along with its main opponents. We then focus on the problem of “Internet piracy” and analyze the efficiency of efforts to reduce the availability of copyrighted creations that are available for non-authorized free download.

National audience

espanolDurante el periodo de vida democratica espanola, hemos ido observando, reiteradamente, innumerables �puzzles� surgidos del resultado de unas elecciones generales, autonomicas o municipales que, como solucion, dan paso a gobiernos compuestos por alianzas, momentaneas o duraderas, de varios partidos politicos. En muchas democracias, es condicion sine qua non, despues de unas elecciones, que los partidos politicos que quieran gobernar sumen un numero de representantes que supere el 50% del total de representantes. Ello provoca un periodo de mayor o menor incertidumbre originado por conocer, en principio, cuales podran ser los pactos y con que partidos se alcanzara un gobierno considerado �estable�. Cuando ninguno de los partidos supera este anhelado porcentaje, existe el riesgo de llegar a la peligrosa situacion en donde, desafortunada e injustamente, podria anteponerse el ansia de poder por encima de lo que realmente podria desear en ciudadano al hacer uso de su derecho al voto. ?Que debemos, pues, hacer para conformar un grupo mas cohesionado y coherente, capaz de trabajar con suficiente grado de logica politica? Nuestra aportacion, basada en la toma de decisiones mediante las tecnicas derivadas de las logicas multivalentes (Zadeh,1965), permite hallar la solucion que maximiza el nivel de confianza de la alianza. Esta solucion no solamente toma en cuenta cada grupo politico y su disciplina interna, sino tambien la personalidad, la mentalidad, la tendencia, los valores y los ideales de cada uno de los parlamentarios, asi como su predisposicion a que uno u otro partido controle el gobierno. Como caso practico, hemos realizado un analisis detallado de un ayuntamiento en el que se toman en cuenta los principales temas de interes, divergiendo la valoracion que sobre estos tiene cada concejal segun su ideologia politica. EnglishDuring the life span of Spanish democracy, we have repeatedly observed many puzzles arising from the results of a general, state or municipal election, which, in turn, give way to governments based in partnerships of several political parties. In many democracies, it is a sine qua non condition that, after an election, the political parties that want to form government must have more than 50% of all representatives. This leads to a period of greater or lesser uncertainty caused, in principle, by unknown pacts with parties that can form a government considered stable. When none of the parties exceeds the majority level, there is the risk of the dangerous situation where, unfortunately and unfairly, the thirst for power could take precedence over what the citizens wished when they exercised their right to vote. Therefore, what we must do to form a more cohesive and consistent group, able to work with a sufficient degree of political logic? Our contribution, based on decision-making techniques derived from multivalued logics (Zadeh, 1965), is able to find the solution that maximizes the confidence level of the alliance. This solution not only takes into account each group and its internal discipline, but also the personality, attitudes, trends, values and ideals of individual parliamentarians, as well as their willingness join to one or another party which wants to control the government. As a case study, we performed a detailed analysis of a municipality, taking into account the main themes and diverting valuation we get for each councilman according to their political ideology.

In spite (or perhaps due) to its simplicity, the question of simulating the M/M/1 queue has attracted much interest. It has served as a benchmark as various properties related to the simulation, or to the simulated performance measures, are known for this queue. In this paper we report on some exper

Remote labs offer many unique advantages to students as they provide opportunities to access experiments and learning scenarios that would be otherwise unavailable. At the same time, however, these opportunities introduce real challenges to the institutions hosting the remote labs. This paper draws on the experiences of the REXNET project consortium to expose a number of these issues as a means of furthering the debate on the value of remote labs and the best practices in deploying them. The paper presents a brief outline of the various types of remote lab scenarios that might be deployed. It then describes the key human and technological actors that have an interest in or are intrinsic to a remote lab instance, with a description of the role of each actor and their interest. Some relationships between these various actors are then discussed with some factors that might influence those relationships. Finally some general issues are briefly described.

With the rapid development of short-range wireless technology new venues to apply it in more sophisticated, complex, and dynamic environments have been opened. Nevertheless, the applicability of such technology in non-traditional settings like face-to-face encounters and disaster relief environments, remains unclear. This article describes a research effort aimed to narrow that gap by means of using two non-traditional settings as case studies; face-to-face encounters among unacquainted people and first responders in urban disaster relief environments. Among the results obtained are: a) interactions among unacquainted people may be promoted, though the level of interaction becomes easily constrained due to the current state of RF technology and the design of the experiments, and b) it is feasible to obtain a reliable communication platform for first responders operating in disaster relief missions. These results supports the idea that short-range wireless technology may play both a facilitator and a promoter role in face-to-face contexts, and at least a facilitator role in the case of users co-located in highly dynamic contexts.

Remote experimentation laboratories are systems based on real equipment, allowing students to perform practical work through a computer connected to the internet. In engineering fields lab activities play a fundamental role. Distance learning has not demonstrated good results in engineering fields because traditional lab activities cannot be covered by this paradigm. These activities can be set for one or for a group of students who work from different locations. All these configurations lead to considering a flexible model that covers all possibilities (for an individual or a group). An inter-continental network of remote laboratories supported by both European and Latin American institutions of higher education has been formed. In this network context, a learning collaborative model for students working from different locations has been defined. The first considerations are presented. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Los sistemas colaborativos pertenecen al área de investigación llamada CSCW (“Computer-Supported Cooperative Work”). Estos sistemas apoyan a grupos de personas trabajando en equipo, con el fin de alcanzar metas comunes. Un aspecto fundamental en todo proceso de colaboración es la comunicación, y aunque todos los sistemas siguen el mismo patrón arquitectónico, la falta de una definición explícita del patrón hace que sea imposible el reuso de soluciones en este ámbito. Es por eso que en el presente artículo se presenta el patrón arquitectónico U/PC (Users/Processes Communication) que modela los mecanismos de comunicación en los sistemas colaborativos, brindando una solución al problema antes mencionado.

Los sistemas colaborativos pertenecen al área de investigación llamada CSCW ("ComputerSupported Cooperative Work"). Estos sistemas apoyan a grupos de personas trabajando en equipo, con el fin de alcanzar metas comunes. Un aspecto fundamental en todo proceso de colaboración es la comunicación. En el presente artículo se propone un esquema de comunicación que permite modelar los mecanismos de distribución de mensajes que se requieren en todos los sistemas colaborativos. Estos mecanismos proveen comunicación tanto entre usuarios como entre procesos del sistema a través del cual se está colaborando.