Maestría en Ciencias de la Información y las Comunicaciones
URI permanente para esta colecciónhttp://hdl.handle.net/11349/18
Examinar
Envíos recientes
Ítem Como una arquitectura de microservicios apoya el despliegue de terminalesNariño Nino, Cristian; Salcedo Parra, Octavio Jose; Narino Nino, Cristian [0000-0002-9440-1045]; Salcedo Parra, Octavio Jose [0000-0002-0767-8522]In recent years, dataphone payment processing networks have been living with the problem of updating dataphone software in the Field, which generates significant operating costs. In addition, the time your application market with the great changes in the payment market has lagged behind due to the speed of its developments. One of the problems that appear in this scenario is the loss of competitiveness of the payment processing networks and this increases if the business processes are not aligned and integrated with the organization's strategy. This logistics operation situation has been creating the need for a new architectural approach to structured, integrated and microservices POS software with a business vision whose application allows payment processing networks to be competitive with the new payment products in the POS and reduce the operating costs of having to visit the street every time you do a software update. This microservices architecture arises as a necessity for means of payment companies that need to reach the field remotely as well as optimize new application development times. Today microservices architectures provide software development advantages that are necessary in POS software models. The main advantages sought are the following: Modularity, Scalability, Versatility, Speed of action Simple and cheap maintenance, Agility.Ítem Pronóstico de contaminación por material particulado, mediante redes neuronales artificiales para generar un sistema de alarmas ambientalesCorzo Muñoz, Miguel Angel; Salcedo Parra, Octavio José; Salcedo Parra, Octavio José [0000-0002-0767-8522 ]Through this project, a web application is developed that warns users about the risk of going out without respiratory protection, according to the forecast of pollution levels for the next 72 hours, obtained through Artificial Neural Networks (ANN), whose input data are the previous measurements from the air quality station installed in that area.Ítem Desarrollo de software orientado a dispositivos móviles basado en realidad aumentada para la validación de sistemas de información geográfica a nivel catastralRamírez Navarro, Sergio Alexander; Barragán Zaque, WilliamThe augmented reality in mobile devices is a concept that over the years has been taking a significant value increasingly higher, this added to the development of geographic information systems and the cadastre, has led to a growth of the same going from exclusive research rooms, to be part of the common use, and soon, to the specific use in different areas of scientific knowledge. This research presents the development of software oriented to mobile devices based on augmented reality systems for the validation of geographic information systems at cadastral level, showing all the steps for its development, the difficulties to reach the solution and even the proposition of photogrammetric concepts to reach the final result. The development of the CatSIGAR application was finally achieved, which handles geographic information and augmented reality in Android mobile devices, establishing starting points for a deeper development of this topic, which is at an important point to find a new solution for the management of geographic information at the cadastral level.Ítem Clasificación de eventos astronómicos transitorios con redes neuronales recurrentesMartin Vega, Daniela; Salcedo Parra, Octavio José; Chaparro Molano, Germán; Salcedo Parra, Octavio José [0000-0002-0767-8522]In this master thesis, an innovative methodology based on recurrent neural networks (RNN) is introduced to address the accurate detection and classification of transient astronomical events from real observational data. Based on the inherent ability of RNNs to model data sequences and adapt to the temporal variability of events in space, the approach employs the MANTRA dataset, covering a diversity of events such as supernovae and active galaxy nuclei. Two preprocessing architectures are developed and gated recurrent units (GRUs) are implemented in a specialized recurrent neural network. Experimental results reveal a remarkable performance improvement, with up to 17% increase in accuracy, highlighting the training efficiency compared to conventional machine learning approaches. This advance contributes significantly to the automation of astronomical event classification, facilitating the early detection of astrophysical phenomena and enriching our understanding of the ever-changing universe.Ítem Determinación del factor de cambio en el glaciar de la Sierra Nevada de Santa Marta, a partir de regiones en movimiento en bases de datos espacio temporales.Porras Martin, Camilo Andres; Ortiz Dávila, Alvaro Enrique; Ortiz Dávila Alvaro Enrique [0000-0001-8830-1657]The evolution of databases over the years, with the aim of storing and managing large volumes of information through computerized systems (Durango Vanegas, 2013), has allowed the development of new types of databases, such as spatial databases, which allow the management of geographic data, its associated data, and the design data of the geometries, assisted by computational intelligence (R. H. Güting, 2005), these databases improved the management of spatial information giving step to geographic information systems, however time is a dimension that allows us to understand and perform analysis of spatial data in a better way, thus producing a relationship between spatiality and time that later came to be called spatial-temporal data stored in data vector, thus achieving the construction of spatio-temporal databases that approximates They are increasingly up to date and deliver data and analysis more accurately.Ítem Diseño y validación de una arquitectura de software para sistemas de predicción bursátil basados en aprendizaje de máquinaRomero Ramirez, Johan Alfredo; Aparicio Pico, Lilia Edith; Aparicio Pico, Lilia Edith [0000-0003-1841-4423]In this investigation, a software architecture model is proposed to handle multiple Machine Learning algorithmic strategies to generate predictions over the stock market .The proposal were design under the Design Research in Information Systems as the methodological framework and resulted in several software UML artifacts such as dynamic and structural models. As part of the analysis, a functional prototype was implemented and the architecture was validated. As a result, the prototype fulfilled the adaptability requirement and some performance issues were acknowledged along with some opportunities for improvements and future work.Ítem Un modelo de regresión logística para el estudio de la relocalización industrial en Colombia del periodo 2005 - 2015Mesa Ospina, Andrés David; Fuentes López, Héctor Javier; Fuentes López, Héctor Javier [0000-0001-6899-4564]Analysis of the industrial relocation for Colombia from 2005 to 2015. This exercise is based on the information provided by the superintendence of companies. With this information the direction of the industry can be tracked, and that allow to identify if an industry has changed or remained in the same location it had in 2005. Graphical outputs are made to represent the phenomenon and an econometric model that allows determining which variables may have influenced the changes in its location for Colombia, as well as for its main industrial cities. It is found that within the main variables that increase the probability of relocation, they correspond to the planning instruments, population, land prices, and small size of the industry.Ítem Implementación de la técnica multicriterio codas para la toma de decisiones en redes de radio cognitivasVaca Villamil, Tania Alexandra; Hernández Suarez, Cesar Augusto; Giral Ramírez, Diego Armando; Hernández Suarez, Cesar Augusto [0000-0001-9409-8341]; Giral Ramírez, Diego Armando [0000-0001-9983-4555]There are multiple investigations on the occupation of the radioelectric spectrum through Cognitive Radio (CR). Decision making is a key aspect in CR to improve QoS indicators. The decision-making process has multiple variables to analyze, MCDM-based algorithms are widely used in this type of problem due to their efficient results and low computational load. The MCDM algorithm implemented is Combinative Distance-based Assessment (CODAS), the metrics obtained in CODAS are compared with Simple Additive Weighting (SAW) and with a RANDOM selection of them, as input information spectral power measurements are used. To establish the performance of the algorithm, five QoS metrics are used: Number Handoffs, Number Failed Handoffs, Average Bandwidth, Average Throughput, Cumulative Average Delay. According to the results obtained, CODAS presented the best result, for the cost metrics with the lowest levels, for the benefit metrics the highest levels were obtainedÍtem Modelo de sistemas multi-agentes para la consulta basada en niveles de confianza, sobre recursos vinculados mediante datos abiertos enlazadosValencia Martínez, Camilo Alejandro; Gaona Garcia, Paulo Alonso; Gaona Garcia, Paulo Alonso [0000-0002-8758-1412]This degree project has the purpose of contributing to the Semantic Web area, specifically on the linking of open data LOD. The foregoing through the development of a multi-agent model that allows for a semantic analysis of the current state of open digital resources based on principles defined by Linked Open Data (LOD) (Berners-Lee, Hendler, Lassila, & others, 2001). The proposed model would be supported by an ontology that allows defining the levels of confidence in open resources, supported by proposals that manage levels of confidence in digital resources, and Semantic Web tools, such as reasoners, ontology processors, and graph processing.Ítem Modelo de computación en la nube para el telemonitoreo de signos vitales en tiempo realDaza Rojas, John Jairo; Aparicio Pico, Lilia Edith; Aparicio Pico Lilia Edith [0000-0003-1841-4423]The book content initially addressed a study of the state of the art of current vital signs telemonitoring solutions, based on the results obtained, a solution design based on cloud computing is proposed, which offers a digital space that integrates everything Necessary to provide a solution to the needs of the health service to patients, the design responds to the needs of health care and life sciences that seek to aggregate data remotely and from multiple sources to determine the condition of a patient. A software prototype for telemonitoring of vital signs was developed that uses aggregated data leading to a better understanding of the patient's condition offering the ability to model the progression of the underlying condition, eventually providing the potential to predict the onset of symptoms. conditions. The thesis work lays the foundations to continue exploring new mechanisms to extract data from medical devices, mobile applications and even chips integrated into the human body.Ítem Metodología para la publicación estandarizada de firmas espectrales en la clasificación de coberturas del sueloValbuena Gaona, Martha Patricia; Herrera Escorcia, José LuisSpectral signatures are graphs that show the reflectance of objects in the presence of light. There are particularities in the shape of the spectral curve due to the physical and chemical composition of the different soil coverings. Therefore, the inclusion of spectral signatures in object-based classification represents a determining factor in the accuracy of classifications. The definition of a standard for the publication of spectral signatures represents an advance in the field of remote sensing as it facilitates researchers to access the results of other studies, verifying their quality and accuracy, making possible the formation of a national library of spectral signatures. Through this project, the development of a methodology to standardize the publication of spectral signatures in the classification of soil coverings is proposed, taking into account the international standards proposed by the OGC, through the Sensors Web Enabling Framework - SWE, which allows developers to make information captured by different instruments detectable, accessible and usable via the Web. The research will be carried out in three stages, the first includes the bibliographic review and regulations related to the standardization of spectral signatures at the national and international levels. Subsequently, a standard metadata is defined based on the OGC parameters, identifying the representative parameter settings required for the classification of soil coverages. Finally, the proposed methodology requires field validation, where fields defined in the metadata shall be verified to be representative from Machine Learning methods.Ítem Metodología para la aplicación de la ciencia de datos en el diagnóstico del cáncer de mamaMillán Gómez, Jorge Armando; Aparicio Pico, Lilia EdithIn 2020 the detected cases of breast cancer in Colombia were 15,509 of which 4,411 ended in death. The anticipate prognosis of this disease has become a research need because it can facilitate preventive treatment to avoid its lethality in an advanced stage. This paper proposes the DSM-BCD methodology (Data science methodology for breast cancer diagnosis) designed to speed up the diagnosis of breast cancer through the continuous improvement of Machine Learning and Deep Learning techniques based on the insight of the oncology specialist and the feedback of knowledge according to the behavior of the data in the various techniques for the detection of breast cancer.Ítem Metodología para identificar frailejón mediante el uso de imágenes de radar: una herramienta para la conservación ambientalSandoval Lagos, Jonathan Said; Castillo Méndez , Luis EduardoOne of the main characteristics of the images captured by synthetic aperture radars (SAR) is the possibility of being generated in almost any climatic condition, thus resulting in extracting information from the earth's surface in areas with a predominant presence of clouds, such as the ecosystem of Paramo. In this work, a methodology for the detection of frailejon was developed, evaluated and validated through the capture of information in the field and the use of Sentinel radar images over an area located in the Sumapaz paramo. Unsupervised and supervised classification methods were used, obtaining the best results in the evaluation phase with the Support Vector Machine (SVM) technique, obtaining a kappa coefficient of 0.76 and an overall accuracy of 88%. Subsequently, as part of the evaluation process, the methodology was replicated on a new area located in the Sumapaz páramo using the same image used in the classification described above and obtaining as a result a kappa coefficient of 0.78 and a global accuracy of 89. % . Lastly, the proposed methodology was validated by applying the same classification method to an image with a different temporality than the one used in the evaluation process, obtaining a kappa coefficient of 0.82 and a global accuracy of 92%. Therefore, the implementation of the methodology reveals the great potential of involving images from active sensors in the identification of frailejon.Ítem Metodología para la predicción de problemas referentes a las TIC en organizaciones de servicios generalesPinzón, Christian Jesus; Vega Escobar, Adriana Marcela; Vega Escobar, Adriana Marcela [0000-0003-4739-2606]This thesis is to show the efficiency of implementing a help desk application with a predictive model for the definition of a catalog of services, with the execution of good practices for the management of technological services through a quality standard to guarantee these services. accordingly, the general services company intends to resolve the requests and requirements of users through application, keeping the life cycle of each incident centralized. This Help Desk system was developed using the PHP programming language, with the HTML5 structure, its CSS styles defined by Bootstrap and its MYSQL database engine. With the creation of an entity-relationship model suitable for the needs of the company, a ticket management module, user module, manual module and a descriptive analysis module of the tickets found in the base.Ítem Diseño e implementación de un entorno IoT en cultivos de feijoa, en la vereda de Piedra-larga del municipio de Ciénega – BoyacáGonzalez Prieto, Fredy Alexander; Salcedo Parra, Octavio Jose; Salcedo Parra, Octavio Jose [0000-0002-0767-8522]This document presents the project with an emphasis on deepening the area of telematics, called "Design and Implementation of an IoT Environment in corn crops, in the village of Piedra-larga in the municipality of Ciénega - Boyacá" which seeks to develop a method of supporting the means of agricultural production with a technological development facilitating the care of crops by farmers This paper documents the necessary information for the implementation of a soil variable monitoring system (through IoT) for the control of a drip irrigation system, this system collects the data provided by temperature sensors. , relative humidity and Ph of the cultivated land, implemented in different sectors of the crop in a plot of the village of Piedra Larga in the municipality of Ciénega. This system assists the farmer in the following way: a) The activation of the drip irrigation system seeking the efficient use of water, b) Improve fruit production by controlling the percentage of nutrients to improve feijoa crop production. c) Provide farmers with access to information on the state of the crop. In the development of the system, a mobile application was designed that shows real-time data monitoring of environmental and soil variables, for the analysis of results and the concentrations of the nutrient mixture together with the drip control to be applied to the crop. An optimal estimation of the required nutrient concentrations was made from a neural network to simplify and improve the efficiency of the farmer's agricultural activities, such as savings in water consumption by 40% and improvement in fruit production by up to a 30% Más información sobre este texto de origenPara obtener más información sobre la traducción, se necesita el texto de origen Enviar comentarios Paneles lateralesÍtem Aplicación de machine learning para la orientación en la toma de decisiones frente al uso agrícola apropiado del suelo para zonas con cultivos ilícitos en ColombiaMatta Monroy, Nancy Johana; Vera Parra, Nelson Enrique; Vera Parra, Nelson Enrique [0000-0002-5159-9207]The eradication of crops for illicit use in Colombia is the closest thing to a chimera: according to figures from the Drug Observatory, the source that offers official numbers on this matter, in 2018, 59,977 hectares of coca were eradicated manually, a laudable effort but that, without a doubt, falls short compared to the 169,018 hectares of coca registered that same year in the country. Agriculture in Colombia plays an important role in the economy and employment, although there is a common difficulty among Colombian farmers, which is that they do not choose the right crop based on productivity, climatic conditions, biota and soil properties, therefore productivity and quality are affected. The main objective of the research was to guide decision-making in determining the best agricultural use of the land where there are illicit crops in Colombia, through the application of machine learning, where as the main product an orientation instrument was obtained for growers who want replace illicit crops based on soil properties, climatic and biotic conditions, and productivity. For the development of this, the Knowledge Discovery in Databases KDD process was applied. The study evidenced that the best automatic methods to predict the substitutable crop are those based on Random Forest and that the learning application is an important decision support tool for crop prediction.Ítem Sistema de identificación y análisis de variables de riesgo para el cálculo de primas de reaseguro empleando modelos computacionalesVargas Carrillo, Luz Angela; Rodríguez Rodríguez, Jorge EnriqueThe purpose of this document is to build a model based on decision tress and learning algorithms, which will allow to predict, if a business opportunity that arrives to a reinsurance brokerage firm can be won or lost. In the same manner to obtain decision rules which can optimize the internal process of the company, in such a way as to reduce the percentage of loss of business and increase customer’s retention rate.Ítem Modelo de identidad digital para documentos administrativos mediante firma electrónica y BlockChainArcos Muñoz, Noé; Jacinto Gómez, Edwar; Rodríguez Guerrero, Rocío; Jacinto Gómez, Edwar [0000-0003-4038-8137]Nowadays, teleworking has entered private and public institutions as an improvised solution in the development of work activities, and this process implies some difficulties in the veracity of the digital supports obtained. This work proposes to implement the concept of identity in the digital documentation obtained, through an intelligent model that contemplates the combined use of encryption techniques and block chains that allow the authentication and identification of users and documents, taking as reference the integrated management system. of processes that the company or institution has.Ítem Soluciones cercanas al optimo para el problema del ladrón viajero a través de un algoritmo genético paralelo implementado en unidades de procesamiento grafico (gpus)Wisk, Sebastian; Poveda Chaves, Roberto Manuel; Poveda Chaves, Roberto Manuel [0000-0002-6694-7673]The Traveling Thief Problem (TTP) is a new and important combinatorial optimization problem that combines two outstanding problems of the NP-Hard class; which are the Traveling Salesman Problem (TSP) and the Knapsack Problem (KP). The TTP has been attempted to be solved using different algorithms and heuristics; The research proposed in this paper will seek an implementation on GPUs of a parallel genetic algorithm to find solutions close to the optimal through the exhaustive use of multiprocessing hardware and the appropriate use of memory spaces. The problems tested correspond to prominent problems in the literature (Benchmark Problems) and the results of the execution will be compared against other results documented so far to determine the validity of our algorithm.Ítem Modelo de clasificación supervisado para la detección de la Roya en hojas del cultivo de café aplicando técnicas de Machine LearningRocha Calderón, Camilo Enrique; Barón Velandia, Julio; Vanegas Ayala, Sebastian Camilo; Vanegas Ayala, Sebastian Camilo [0000-0002-8610-9765]This proposal focuses on the development of a model that allows the early detection of the Rust disease and its different stages presented in the leaves of the coffee crop, applying Machine Learning techniques characterized by their interpretability, improved with respect to their level of accuracy through hybrid optimization algorithms, providing an alternative to traditional algorithms and models of image analysis. Therefore, the model will stand out for its high performance in the analysis and classification of images, identifying the state of a plant (healthy or diseased), allowing farmers to take corrective measures in a timely manner, reducing the impact generated by the disease on crop yield.