University of Cauca

Machine Learning (ML) has been enjoying an unprecedented surge in applications that solve problems and enable automation in diverse domains. Primarily, this is due to the explosion in the availability of data, significant improvements in ML techniques, and advancement in computing capabilities. Undoubtedly, ML has been applied to various mundane and complex problems arising in network operation and management. There are various surveys on ML for specific areas in networking or for specific network technologies. This survey is original, since it jointly presents the application of diverse ML techniques in various key areas of networking across different network technologies. In this way, readers will benefit from a comprehensive discussion on the different learning paradigms and ML techniques applied to fundamental problems in networking, including traffic prediction, routing and classification, congestion control, resource and fault management, QoS and QoE management, and network security. Furthermore, this survey delineates the limitations, give insights, research challenges and future opportunities to advance ML in networking. Therefore, this is a timely contribution of the implications of ML for networking, that is pushing the barriers of autonomic network operation and management.

Due to the significant advancement of the Internet of Things (IoT) in the healthcare sector, the security, and the integrity of the medical data became big challenges for healthcare services applications. This paper proposes a hybrid security model for securing the diagnostic text data in medical images. The proposed model is developed through integrating either 2-D discrete wavelet transform 1 level (2D-DWT-1L) or 2-D discrete wavelet transform 2 level (2D-DWT-2L) steganography technique with a proposed hybrid encryption scheme. The proposed hybrid encryption schema is built using a combination of Advanced Encryption Standard, and Rivest, Shamir, and Adleman algorithms. The proposed model starts by encrypting the secret data; then it hides the result in a cover image using 2D-DWT-1L or 2D-DWT-2L. Both color and gray-scale images are used as cover images to conceal different text sizes. The performance of the proposed system was evaluated based on six statistical parameters; the peak signal-to-noise ratio (PSNR), mean square error (MSE), bit error rate (BER), structural similarity (SSIM), structural content (SC), and correlation. The PSNR values were relatively varied from 50.59 to 57.44 in case of color images and from 50.52 to 56.09 with the gray scale images. The MSE values varied from 0.12 to 0.57 for the color images and from 0.14 to 0.57 for the gray scale images. The BER values were zero for both images, while SSIM, SC, and correlation values were ones for both images. Compared with the state-of-the-art methods, the proposed model proved its ability to hide the confidential patient's data into a transmitted cover image with high imperceptibility, capacity, and minimal deterioration in the received stego-image.

BACKGROUND: Despite the increasing number of manuals on how to develop clinical practice guidelines (CPGs) there remain concerns about their quality. The aim of this study was to review the quality of CPGs across a wide range of healthcare topics published since 1980. METHODS: The authors conducted a literature search in MEDLINE to identify publications assessing the quality of CPGs with the Appraisal of Guidelines, Research and Evaluation (AGREE) instrument. For the included guidelines in each study, the authors gathered data about the year of publication, institution, country, healthcare topic, AGREE score per domain and overall assessment. RESULTS: In total, 42 reviews were selected, including a total of 626 guidelines, published between 1980 and 2007, with a median of 25 CPGs. The mean scores were acceptable for the domain 'Scope and purpose' (64%; 95% CI 61.9 to 66.4) and 'Clarity and presentation' (60%; 95% CI 57.9 to 61.9), moderate for domain 'Rigour of development' (43%; 95% CI 41.0 to 45.2), and low for the other domains ('Stakeholder involvement' 35%; 95% CI 33.9 to 37.5, 'Editorial independence' 30%; 95% CI 27.9 to 32.3, and 'Applicability' 22%; 95% CI 20.4 to 23.9). From those guidelines that included an overall assessment, 62% (168/270) were recommended or recommended with provisos. There was a significant improvement over time for all domains, except for 'Editorial independence.' CONCLUSIONS: This review shows that despite some increase in quality of CPGs over time, the quality scores as measured with the AGREE Instrument have remained moderate to low over the last two decades. This finding urges guideline developers to continue improving the quality of their products. International collaboration could help increasing the efficiency of the process.

El presente libro recoge dos ensayos escritos entre el 2011 y el 2013, y una breve propuesta de investigación hacia el futuro que surge de los mismos. Los textos constituyen nuevas lecturas sobre desarrollo, territorio y diferencia, y contribuyen a delinear un campo que provisionalmente denominamos como "ontología política".\nComo veremos en el segundo ensayo, el concepto de ontología política busca resaltar tanto la dimensión política de la ontología como la dimensión ontológica de la política. Por un lado, toda ontología o visión del mundo crea una forma particular de ver y hacer la política; por el otro, muchos conflictos políticos nos refieren a premisas fundamentales sobre lo que son el mundo, lo real y la vida; es decir, a ontologías.

Recommender systems have been based on context and content, and now the technological challenge of making personalized recommendations based on the user emotional state arises through physiological signals that are obtained from devices or sensors. This paper applies the deep learning approach using a deep convolutional neural network on a dataset of physiological signals (electrocardiogram and galvanic skin response), in this case, the AMIGOS dataset. The detection of emotions is done by correlating these physiological signals with the data of arousal and valence of this dataset, to classify the affective state of a person. In addition, an application for emotion recognition based on classic machine learning algorithms is proposed to extract the features of physiological signals in the domain of time, frequency, and non-linear. This application uses a convolutional neural network for the automatic feature extraction of the physiological signals, and through fully connected network layers, the emotion prediction is made. The experimental results on the AMIGOS dataset show that the method proposed in this paper achieves a better precision of the classification of the emotional states, in comparison with the originally obtained by the authors of this dataset.

The aim of this study was to evaluate ultrasonic P-wave velocity as a feature for predicting some physical and mechanical properties that describe the behavior of local building limestone.To this end, both ultrasonic testing and compressive tests were carried out on several limestone specimens and statistical correlation between ultrasonic velocity and density, compressive strength, and modulus of elasticity was studied.The effectiveness of ultrasonic velocity was evaluated by regression, with the aim of observing the coefficient of determination 2 between ultrasonic velocity and the aforementioned parameters, and the mathematical expressions of the correlations were found and discussed.The strong relations that were established between ultrasonic velocity and limestone properties indicate that these parameters can be reasonably estimated by means of this nondestructive parameter.This may be of great value in a preliminary phase of the diagnosis and inspection of stone masonry conditions, especially when the possibility of sampling material cores is reduced.

In this work, a methodology of synthesis was designed to obtain ZnO nanoparticles (ZnO NPs) in a controlled and reproducible manner. The nanoparticles obtained were characterized using infrared spectroscopy, X-ray diffraction, and transmission electron microscopy (TEM). Also, we determined the antifungal capacity in vitro of zinc oxide nanoparticles synthesized, examining their action on Erythricium salmonicolor fungy causal of pink disease. To determine the effect of the quantity of zinc precursor used during ZnO NPs synthesis on the antifungal capacity, 0.1 and 0.15 M concentrations of zinc acetate were examined. To study the inactivation of the mycelial growth of the fungus, different concentrations of ZnO NPs of the two types of synthesized samples were used. The inhibitory effect on the growth of the fungus was determined by measuring the growth area as a function of time. The morphological change was observed with high-resolution optical microscopy (HROM), while TEM was used to observe changes in its ultrastructure. The results showed that a concentration of 9 mmol L−1 for the sample obtained from the 0.15 M and at 12 mmol L−1 for the 0.1 M system significantly inhibited growth of E. salmonicolor. In the HROM images a deformation was observed in the growth pattern: notable thinning of the fibers of the hyphae and a clumping tendency. The TEM images showed a liquefaction of the cytoplasmic content, making it less electron-dense, with the presence of a number of vacuoles and significant detachment of the cell wall.

Over the last decade, a significant amount of effort has been invested on architecting agile and adaptive management solutions in support of autonomic, self-managing networks. Autonomic networking calls for automated decisions for management actions. This can be realized through a set of pre-defined network management policies engineered from human expert knowledge. However, engineering sufficiently accurate knowledge considering the high complexity of today's networking environment is a difficult task. This has been a particularly limiting factor in the practical deployment of autonomic systems. ML is a powerful technique for extracting knowledge from data. However, there has been little evidence of its application in realizing practical management solutions for autonomic networks. Recent advances in network softwarization and programmability through SDN and NFV, the proliferation of new sources of data, and the availability of lowcost and seemingly infinite storage and compute resource from the cloud are paving the way for the adoption of ML to realize cognitive network management in support of autonomic networking. This article is intended to stimulate thought and foster discussion on how to defeat the bottlenecks that are limiting the wide deployment of autonomic systems, and the role that ML can play in this regard.

The most common cause of acquired thyroid dysfunction is autoimmune thyroid disease, which is an organ-specific autoimmune disease with two presentation phenotypes: hyperthyroidism (Graves-Basedow disease) and hypothyroidism (Hashimoto's thyroiditis). Hashimoto's thyroiditis is distinguished by the presence of autoantibodies against thyroid peroxidase and thyroglobulin. Meanwhile, autoantibodies against the TSH receptor have been found in Graves-Basedow disease. Numerous susceptibility genes, as well as epigenetic and environmental factors, contribute to the pathogenesis of both diseases. This review summarizes the most common genetic, epigenetic, and environmental mechanisms involved in autoimmune thyroid disease.

Pentavalent antimony, the generally accepted treatment for leishmaniasis, is given parenterally, and it is expensive and not readily available In developing countries. An inexpensive, orally administered compound would be a substantial advance in treatment. Previous studies in vitro have shown synergism between allopurinol and pentavalent antimony in tissue-culture systems. We designed this clinical study to determine whether synergism could be demonstrated In patients.

n recent years it has been shown that the secure exchange of medical information significantly benefits people’s life quality, improving their care and treatment. The interoperability of the entire healthcare ecosystem is a constant challenge, and even more, with all the risks posed to the security of healthcare information. Blockchain technology is emerging as one of the main alternatives when it comes to finding a balance in the healthcare ecosystem. However, the constant development of new Blockchain technologies and the evolution of healthcare systems make it difficult to find established proposals. From an architectural point of view, the design of blockchain-based solutions requires trade-offs e.g., security and interoperability. This paper focuses on two main objectives, in the first one, it was carried out a Systematic Literature Review for exploring architectural mechanisms used to support the interoperability and security of Blockchain-based Health Management Systems. Taking into account of results, a series of scenarios were generated where these mechanisms can be used along with their context, issues, and various architectural concerns (interoperability and security). In the second objective, a high-level architecture and its validation were proposed through an experiment for the whole process of developing a Domain Specific Language, using the Model Driven Engineering methodology for specific Smart Contracts.

As global demand for livestock products (such as meat, milk and eggs) is expected to double by 2050, necessary increases to future production must be reconciled with negative environmental impacts that livestock cause. This paper describes the LivestockPlus concept and demonstrates how the sowing of improved forages can lead to the sustainable intensification of mixed crop-forage-livestock-tree systems in the tropics by producing multiple social, economic and environmental benefits. Sustainable intensification not only improves the productivity of tropical forage-based systems but also reduces the ecological footprint of livestock production and generates a diversity of ecosystem services (ES) such as improved soil quality and reduced erosion, sedimentation and greenhouse gas (GHG) emissions. Integrating improved grass and legume forages into mixed production systems (crop-livestock, tree-livestock, crop-tree-livestock) can restore degraded lands and enhance system resilience to drought and waterlogging associated with climate change. When properly managed tropical forages accumulate large amounts of carbon in soil, fix atmospheric nitrogen (legumes), inhibit nitrification in soil and reduce nitrous oxide emissions (grasses), and reduce GHG emissions per unit livestock product.The LivestockPlus concept is defined as the sustainable intensification of forage-based systems, which is based on 3 interrelated intensification processes: genetic intensification - the development and use of superior grass and legume cultivars for increased livestock productivity; ecological intensification - the development and application of improved farm and natural resource management practices; and socio-economic intensification - the improvement of local and national institutions and policies, which enable refinements of technologies and support their enduring use. Increases in livestock productivity will require coordinated efforts to develop supportive government, non-government organization and private sector policies that foster investments and fair market compensation for both the products and ES provided. Effective research-for-development efforts that promote agricultural and environmental benefits of forage-based systems can contribute towards implemention of LivestockPlus across a variety of geographic, political and socio-economic contexts.Keywords: Eco-efficiency, environmental benefits, livestock and environment, mixed farming, pastures, smallholders.DOI: 10.17138/TGFT(3)59-82

Traditional routing protocols employ limited information to make routing decisions, which can lead to a slow adaptation to traffic variability, as well as restricted support to the Quality of Service (QoS) requirements of applications. This article introduces a novel approach for routing in Software-defined networking (SDN), called Reinforcement Learning and Software-Defined Networking Intelligent Routing (RSIR). RSIR adds a Knowledge Plane to SDN and defines a routing algorithm based on Reinforcement Learning (RL) that takes into account link-state information to make routing decisions. This algorithm capitalizes on the interaction with the environment, the intelligence provided by RL and the global view and control of the network furnished by SDN, to compute and install, in advance, optimal routes in the forwarding devices. RSIR was extensively evaluated by emulation using real traffic matrices. Results show RSIR outperforms the Dijkstra's algorithm in relation to the stretch, link throughput, packet loss, and delay when available bandwidth, delay, and loss are considered individually or jointly for the computation of optimal paths. The results demonstrate that RSIR is an attractive solution for intelligent routing in SDN.

BACKGROUND: Long chain polyunsaturated fatty acids (LCPUFA), especially docosahexaenoic acid (DHA), are the most abundant fatty acids in the brain and are necessary for growth and maturation of an infant's brain and retina. LCPUFAs are named "essential" because they cannot be synthesised efficiently by the human body and come from maternal diet. It remains controversial whether LCPUFA supplementation to breastfeeding mothers is beneficial for the development of their infants. OBJECTIVES: To assess the effectiveness and safety of supplementation with LCPUFA in breastfeeding mothers in the cognitive and physical development of their infants as well as safety for the mother and infant. SEARCH METHODS: We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (6 August 2014), CENTRAL (Cochrane Library 2014, Issue 8), PubMed (1966 to August 2014), EMBASE (1974 to August 2014), LILACS (1982 to August 2014), Google Scholar (August 2014) and reference lists of published narrative and systematic reviews. SELECTION CRITERIA: Randomised controlled trials or cluster-randomised controlled trials evaluating the effects of LCPUFA supplementation on breastfeeding mothers (including the pregnancy period) and their infants. DATA COLLECTION AND ANALYSIS: Two review authors independently assessed eligibility and trial quality, performed data extraction and evaluated data accuracy. MAIN RESULTS: We included eight randomised controlled trials involving 1567 women. All the studies were performed in high-income countries. The longest follow-up was seven years.We report the results from the longest follow-up time point from included studies. Overall, there was moderate quality evidence as assessed using the GRADE approach from these studies for the following outcomes measured beyond 24 months age of children: language development and child weight. There was low-quality evidence for the outcomes: Intelligence or solving problems ability, psychomotor development, child attention, and child visual acuity.We found no significant difference in children's neurodevelopment at long-term follow-up beyond 24 months: language development (standardised mean difference (SMD) -0.27, 95% confidence interval (CI) -0.56 to 0.02; two trials, 187 participants); intelligence or problem-solving ability (three trials, 238 participants; SMD 0.00, 95% CI -0.36 to 0.36); psychomotor development (SMD -0.11, 95% CI -0.48 to 0.26; one trial, 113 participants); motor development (SMD -0.23, 95% CI -0.60 to 0.14; one trial, 115 participants), or in general movements (risk ratio, RR, 1.12, 95% CI 0.58 to 2.14; one trial, 77 participants; at 12 weeks of life). However, child attention scores were better at five years of age in the group of children whose mothers had received supplementation with fatty acids (mean difference (MD) 4.70, 95% CI 1.30 to 8.10; one study, 110 participants)). In working memory and inhibitory control, we found no significant difference (MD -0.02 95% CI -0.07 to 0.03 one trial, 63 participants); the neurological optimality score did not present any difference (P value: 0.55).For child visual acuity, there was no significant difference (SMD 0.33, 95% CI -0.04 to 0.71; one trial, 111 participants).For growth, there were no significant differences in length (MD -0.39 cm, 95% CI -1.37 to 0.60; four trials, 441 participants), weight (MD 0.13 kg, 95% CI -0.49 to 0.74; four trials, 441 participants), and head circumference (MD 0.15 cm, 95% CI -0.27 to 0.58; three trials, 298 participants). Child fat mass and fat mass distribution did not differ between the intervention and control group (MD 2.10, 95% CI -0.48 to 4.68; one trial, 115 participants, MD -0.50, 95% CI -1.69 to 0.69; one trial, 165 participants, respectively).One study (117 infants) reported a significant difference in infant allergy at short-term follow-up (risk ratio (RR) 0.13, 95% CI 0.02 to 0.95), but not at medium-term follow-up (RR 0.52, 95% CI 0.17 to 1.59).We found no significant difference in two trials evaluating postpartum depression. Data were not possible to be pooled due to differences in the describing of the outcome. One study (89 women) did not find any significant difference between the LCPUFA supplementation and the control group at four weeks postpartum (MD 1.00, 95%CI -1.72 to 3.72).No adverse effects were reported. AUTHORS' CONCLUSIONS: Based on the available evidence, LCPUFA supplementation did not appear to improve children's neurodevelopment, visual acuity or growth. In child attention at five years of age, weak evidence was found (one study) favouring the supplementation. Currently, there is inconclusive evidence to support or refute the practice of giving LCPUFA supplementation to breastfeeding mothers in order to improve neurodevelopment or visual acuity.

The online telemedicine systems are helpful since they provide timely and effective healthcare services. Such online healthcare systems are usually based on sophisticated and advanced wearable and wireless sensor technologies. A rapid technological growth has improved the scope of many remote health monitoring systems. Here, the researchers employed a cloud-based remote monitoring system for observing the health status of the patients after monitoring their heart rate variability. This system was developed after considering many factors like the ease of application, costs, accuracy, and the data security. Furthermore, this system was also conceptualized to act as an interface between the patients and the healthcare providers, thus ensuring a two-way communication between them. The major aim of this paper was to provide the best healthcare monitoring services to the people living in the remote areas, which was otherwise very difficult owing to the small doctor-to-patient ratio. The researchers also analyzed their monitoring system using two different databases. First comes from MIT Physionet database i.e., the MIT-BIH sinus rhythm and the MIT-St. Petersburg. While the second database was collected after monitoring 30 people who were asked to use these wearable sensors. After analyzing the performance of the proposed scheme, the obtained results for accuracy, sensitivity, and specificity were 99.02%, 98.78%, and 99.17%, respectively. The achieved results concluded that the proposed system was quite reliable, robust, and valuable. Also, the data analysis revealed that this system was very convenient and ensured data security. In addition, this developed monitoring system generated warning messages, directed towards the patients and the doctors, during some critical situation.

Abstract Soil‐dwelling ants, many of which are generalist predators, are more diverse in shaded than in sun coffee plantations without trees. We compared ant predation on the coffee berry borer, Hypothenemus hampei (Ferrari) (Coleoptera: Curculionidae: Scolytinae) in three shaded and three sun coffee plantations in Apía, Colombia, in both the wet and the dry seasons. We found that H. hampei adults exposed to ants for 5 days suffered higher removal in shaded plantations and in the wet season. In the laboratory, we observed that ants killed 74–99% of H. hampei adults over the course of 5 days. Ants appear to be important predators of H. hampei , particularly in shaded coffee plantations and in the wet season.

Network Slicing and Deep Reinforcement Learning (DRL) are vital enablers for achieving 5G and 6G networks. A 5G/6G network can comprise various network slices from unique or multiple tenants. Network providers need to perform intelligent and efficient resource management to offer slices that meet the quality of service and quality of experience requirements of 5G/6G use cases. Resource management is far from being a straightforward task. This task demands complex and dynamic mechanisms to control admission and allocate, schedule, and orchestrate resources. Intelligent and effective resource management needs to predict the services' demand coming from tenants (each tenant with multiple network slice requests) and achieve autonomous behavior of slices. This paper identifies the relevant phases for resource management in network slicing and analyzes approaches using reinforcement learning (RL) and DRL algorithms for realizing each phase autonomously. We analyze the approaches according to the optimization objective, the network focus (core, radio access, edge, and end-to-end network), the space of states, the space of actions, the algorithms, the structure of deep neural networks, the exploration-exploitation method, and the use cases (or vertical applications). We also provide research directions related to RL/DRL-based network slice resource management.

Abstract In Latin America, the cultivation of Arabica coffee ( Coffea arabica ) plays a critical role in rural livelihoods, biodiversity conservation, and sustainable development. Over the last 20 years, coffee farms and landscapes across the region have undergone rapid and profound biophysical changes in response to low coffee prices, changing climatic conditions, severe plant pathogen outbreaks, and other drivers. Although these biophysical transformations are pervasive and affect millions of rural livelihoods, there is limited information on the types, location, and extent of landscape changes and their socioeconomic and ecological consequences. Here we review the state of knowledge on the ongoing biophysical changes in coffee-growing regions, explore the potential socioeconomic and ecological impacts of these changes, and highlight key research gaps. We identify seven major land-use trends which are affecting the sustainability of coffee-growing regions across Latin America in different ways. These trends include (1) the widespread shift to disease-resistant cultivars, (2) the conventional intensification of coffee management with greater planting densities, greater use of agrochemicals and less shade, (3) the conversion of coffee to other agricultural land uses, (4) the introduction of Robusta coffee ( Coffea canephora ) into areas not previously cultivated with coffee, (5) the expansion of coffee into forested areas, (6) the urbanization of coffee landscapes, and (7) the increase in the area of coffee produced under voluntary sustainability standards. Our review highlights the incomplete and scattered information on the drivers, patterns, and outcomes of biophysical changes in coffee landscapes, and lays out a detailed research agenda to address these research gaps and elucidate the effects of different landscape trajectories on rural livelihoods, biodiversity conservation, and other aspects of sustainable development. A better understanding of the drivers, patterns, and consequences of changes in coffee landscapes is vital for informing the design of policies, programs, and incentives for sustainable coffee production.

Competisoft provides the Latin American software industry with a reference framework for improvement and certification of its software processes. The project is based on proven solutions, including the MoProSoft model that four Mexican software companies applied to increase their processes' capacity level.

El sistema educativo en tiempos de pandemia ha tenido que transformarse de forma urgente e imprevista a una modalidad virtual. En este trabajo se presenta un estudio exploratorio sobre las principales dificultades encontradas por las instituciones educativas en Iberoamérica y algunas estrategias utilizadas en los procesos de enseñanza aprendizaje. Asimismo, teniendo en cuenta el análisis previo, se propone un modelo de evaluación a considerar en los planes de contingencia debido a la emergencia sanitaria.