1,635 results on '"Redes neuronales artificiales"'
Search Results
52. Predicción de efectos fisiológicos causados por el estrés académico mediante redes neuronales artificiales
- Author
-
José Fernando Mora Romo and Juan Martell Muñoz
- Subjects
estrés académico ,procrastinación ,redes neuronales artificiales ,perceptrón multicapa ,psicología ,educación ,Psychology ,BF1-990 - Abstract
Mediante un modelo de perceptrón multicapa (MLP) de redes neuronales artificiales, se buscó realizar un modelo predictivo de efectos fisiológicos causados por el estrés académico. Para esto, se consideran variables como la procrastinación académica, el nivel de estrés percibido en el semestre, el estrés académico, la edad, el ingreso económico familiar e individual. Se obtuvo un porcentaje de pronósticos incorrectos en la fase de prueba y reserva de 38.5% y 19.2%, respectivamente; así como un porcentaje global de clasificación correcto de 80.8% y un valor de área bajo la curva ROC de .752. Las tres variables mayor importancia normalizada dentro del modelo fueron la procrastinación, el nivel de estrés percibido en el semestre y el estrés académico. Por último, se discuten los efectos de la procrastinación y el estrés académico sobre el bienestar físico y psicológico de los estudiantes.
- Published
- 2021
- Full Text
- View/download PDF
53. Selección organizacional: resiliencia y desempeño de las pymes en la era de la COVID-19
- Author
-
Rigoberto García-Contreras, David Valle-Cruz, and Rosa Azalea Canales-García
- Subjects
selección organizacional ,resiliencia organizacional ,desempeño ,covid-19 ,redes neuronales artificiales ,Business ,HF5001-6182 - Abstract
El objetivo del presente artículo fue analizar la perspectiva actual de las pymes ante la crisis de la COVID-19, así como analizar la asociación e incidencia de la resiliencia organizacional en su desempeño durante este periodo crítico. Para probar las hipótesis, se realizó un estudio transversal con una muestra de 112 responsables de pymes en dos países de América Latina (México y Chile). Los métodos utilizados fueron análisis descriptivo de datos, correlación bivariada y redes neuronales artificiales. Los resultados descriptivos demuestran el impacto de la crisis de la COVID-19; además, los resultados prueban la relación e incidencia positiva de la resiliencia en el desempeño de las pymes.
- Published
- 2021
- Full Text
- View/download PDF
54. A neuro-fuzzy inference system for stakeholder classification.
- Author
-
Pérez Vera, Yasiel and Bermudez Peña, Anié
- Subjects
- *
STAKEHOLDERS , *ARTIFICIAL neural networks , *FUZZY logic , *FUZZY neural networks , *PROJECT management , *CLASSIFICATION , *FUZZY systems , *MACHINE learning , *BRAINSTORMING - Abstract
Stakeholder classification is carried out manually using methods such as brainstorming, interviews with experts, and checklists. These methods present a subjective character as they depend on the appreciation of the interviewees. This characteristic affects the accuracy of this classification, making that the project managers do not make the correct decisions. The research aims to suggest a fuzzy inference system for the classification of stakeholders, which will improve the quality of such classification in the projects. The proposal carries out the machine learning and the adjustment of the fuzzy inference system to classify the stakeholders by executing four algorithms based on artificial neural networks: ANFIS, HYFIS, FS.HGD, and FIR.DM. It analyzes the results of applying them in 10 iterations by calculating the measures: percentage of correct classifications, false-positive cases, false-negative cases, and mean square error. The ANFIS system show the best results. The fuzzy inference system for stakeholder classification generated improves the quality of this classification using machine learning, allowing to make better decisions in a project. [ABSTRACT FROM AUTHOR]
- Published
- 2022
55. Fast Facial Landmark Detection and Applications: A Survey.
- Author
-
Khabarlak, Kostiantyn and Koriashkina, Larysa
- Subjects
EMOTION recognition ,HUMAN facial recognition software ,FACE ,COMPUTER vision - Abstract
Copyright of Journal of Computer Science & Technology (JCS&T) is the property of Journal of Computer Science & Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2022
- Full Text
- View/download PDF
56. Análisis de clúster y redes neuronales artificiales en la caracterización y clasificación de perfiles de salud mental positiva en situación de confinamiento por COVID-19.
- Author
-
Navarro-Obeid, Jorge Eduardo, De La Hoz-Granadillo, Efraín Javier, and Vergara-Álvarez, María Laura
- Subjects
ARTIFICIAL neural networks ,SATISFACTION ,CLUSTER analysis (Statistics) ,INTERPERSONAL relations ,MENTAL health - Abstract
Copyright of Gaceta Médica de Caracas is the property of Academia Nacional de Medicina and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2022
- Full Text
- View/download PDF
57. Algoritmo hibrido de redes neuronales artificiales con recocido simulado para predicción en minería de datos
- Author
-
Roberto Emilio Salas ruiz, Jorge Enrique Rodriguez Rodriguez, and Claudia Liliana Hernández García
- Subjects
algoritmos híbridos ,redes neuronales artificiales ,recocido simulado ,predicción de datos ,aprendizaje computacional ,Electronic computers. Computer science ,QA75.5-76.95 ,Computer software ,QA76.75-76.765 - Abstract
El presente artículo es un avance del proyecto de investigación titulado “Desarrollo de algoritmos híbridos para minería de datos” y presenta el uso de una red neuronal con el algoritmo del recocido simulado para realizar la predicción de un conjunto de datos de entrenamiento. En primer lugar, se aborda el problema a resolver, el cual está orientado al análisis de las técnicas definidas para el algoritmo hibrido. Luego, se justifica la metodología de investigación (científica descriptiva-exploratoria con enfoque experimental) aplicada. Se realizó la revisión de las técnicas seleccionadas para la técnica híbrida redes neuronales y recocido simulado la cual se aplica a un conjunto de datos experimentales asociados a determinar en un conjunto de pacientes si su columna vertebral es normal o anormal. Enseguida, se plantean las pruebas de análisis y resultados.
- Published
- 2020
- Full Text
- View/download PDF
58. Optimization of Machining Parameters for Product Quality and Productivity in Turning Process of Aluminum.
- Author
-
Abolghasem, Sepideh and Mancilla-Cubides, Nicolás
- Subjects
- *
PRODUCT quality , *ARTIFICIAL neural networks , *MACHINING , *MANUFACTURING processes , *COGNITIVE processing speed , *SURFACE roughness , *PARTICLE swarm optimization , *SURFACE finishing , *CUTTING tools , *TOOL-steel - Abstract
Modern production process is accompanied with new challenges in reducing the environmental impacts related to machining processes. The turning process is a manufacturing process widely used with numerous applications for creating engineering components. Accordingly, many studies have been conducted in order to optimize the machining parameters and facilitate the decision-making process. This work aims to optimize the quality of the machined products (surface finish) and the productivity rate of the turning manufacturing process. To do so, we use Aluminum as the material test to perform the turning process with cutting speed, feed rate, depth of cut, and nose radius of the cutting tool as our design factors. Product quality is quantified using surface roughness (R_a) and the productivity rate based on material removal rate (MRR). We develop a predictive and optimization model by coupling Artificial Neural Networks (ANN) and the Particle Swarm Optimization (PSO) multi-function optimization technique, as an alternative to predict the model response (R_a) first and then search for the optimal value of turning parameters to minimize the surface roughness (R_a) and maximize the material removal rate (MRR). The results obtained by the proposed models indicate good match between the predicted and experimental values proving that the proposed ANN model is capable to predict the surface roughness accurately. The optimization model PSO has provided a Pareto Front for the optimal solution determining the best machining parameters for minimum R_a and maximum MRR. The results from this study offer application in the real industry where the selection of optimal machining parameters helps to manage two conflicting objectives, which eventually facilitate the decision-making process of machined products [ABSTRACT FROM AUTHOR]
- Published
- 2022
- Full Text
- View/download PDF
59. Social big data y sociología y ciencias sociales computacionales.
- Author
-
Gualda, Estrella
- Subjects
- *
BIG data , *SOCIAL sciences databases , *COMPUTATIONAL sociology , *COVID-19 pandemic , *COMPUTERS in the social sciences , *SOCIAL movements , *ARTIFICIAL intelligence , *MACHINE learning , *ARTIFICIAL neural networks , *DEEP learning , *SOCIAL media - Abstract
The research on what happens on the Internet often brings us closer to studies at the frontier of knowledge. This article is part of the transdisciplinary space of Computational Sociology and Social Sciences. It pretends to present the current development of research referred to as Social Big Data. It describes methodological processes typical in this field where social media on the Internet are the primary data source. Along with this description, it highlights some of the advantages, limitations, and challenges for research in this field, closely linked to methodological and technical advances made in other sciences. The text introduces Social Big Data's conceptual specificity as a confluence of social media, data analysis, and massive data. Then, we explore essential changes in the research process in this field and advances in working with social big data in areas of artificial intelligence such as machine learning, artificial neural networks, and deep learning, which are aligned with social sciences that tend to predictions. We then argue about the relevance for Sociology and other sciences to advance in an approach based on mixed methods in the Social Big Data area, rethinking the micro-macro link in this field of study. Through a case study [on the negationist social movement in Twitter in Spain during the COVID-19 pandemics], we also illustrate some potentials and limitations of this type of research, which will allow us to outline some of the methodological challenges that experts could incorporate into a research agenda in this area. Social Big Data and Computational Sociology and Social Sciences research offer directions of great interest to the Sociology of the coming years in which exceptional progress can be made in the development of transdisciplinarity and hybridization in science, enriching them. [ABSTRACT FROM AUTHOR]
- Published
- 2022
- Full Text
- View/download PDF
60. Variation of surface runoff due to change of land use in the river Duero watershed.
- Author
-
Bernal-Santana, Nelly, Cruz-Cárdenas, Gustavo, Teodoro Silva, José, Martínez-Trinidad, Sergio, Moncayo-Estrada, Rodrigo, Estrada-Godoy, Francisco, Ochoa-Estrada, Salvador, and Álvarez-Bernal, Dioselina
- Subjects
RUNOFF ,LAND use ,TRADITIONAL farming ,DIGITAL mapping ,LAND use mapping ,FOREST conservation ,WATER supply ,FOREST management - Abstract
Copyright of Tecnología y Ciencias del Agua is the property of Instituto Mexicano de Tecnologia del Agua (IMTA) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2022
- Full Text
- View/download PDF
61. Dynamic Stock Dependence and Monetary Variables in the United States (2000-2016): A Copula and Neural Network Approach.
- Author
-
Sosa, Miriam, Bucio, Christian, and Ortiz, Edgar
- Subjects
FOREIGN exchange rates ,SPREAD (Finance) ,ARTIFICIAL neural networks ,STANDARD & Poor's 500 Index ,LIBOR ,MARKET share ,CREDIT derivatives ,INTEREST rates ,STOCK exchanges ,SHORT selling (Securities) - Abstract
Copyright of Lecturas de Economia is the property of Universidad de Antioquia, Facultad de Ciencias Economicas and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2022
- Full Text
- View/download PDF
62. Development of machine learning models for air temperature estimation using MODIS data.
- Author
-
Ovando, G., S., Sayago, and Bocco, M.
- Subjects
ATMOSPHERIC temperature ,LAND surface temperature ,MACHINE learning ,ARTIFICIAL neural networks ,STANDARD deviations ,ATMOSPHERE ,CLIMATE change - Abstract
Copyright of Agriscientia is the property of Revista AgriScientia and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2022
- Full Text
- View/download PDF
63. Predicción de resultados académicos con la aplicación nntool en Matlab utilizando redes neuronales artificiales.
- Author
-
Capuñay Sanchez, Dulce Lucero, Incio Flores, Fernando Alain, Estela Urbina, Ronald Omar, Montenegro Camacho, Luis, Delgado Soto, Jorge Antonio, and Cueva Valdivia, Johnny
- Subjects
ARTIFICIAL neural networks ,STUDENTS ,FOREIGN students ,JUDGMENT (Psychology) ,BASIC education ,STATISTICAL correlation - Abstract
Copyright of Apuntes Universitarios: Revista de Investigación is the property of Universidad Peruana Union Filiar Tarapoto and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2022
- Full Text
- View/download PDF
64. APLICACIÓN DE REDES NEURONALES ARTIFICIALES PARA EL PRONÓSTICO DE PRECIOS DE CAFÉ.
- Author
-
Cardozo Rueda, Karen Shirley
- Published
- 2022
65. Implementación de inteligencia artificial y tecnología blockchain que permita optimizar sistemas productivos de la Pymes.
- Author
-
Acosta Mendizábal, Marco Antonio, Azamar Palma, Iván, De La Rosa Ramírez, Esteban, Pérez Sánchez, Johany Yuliet, Solares Sánchez, Abril, and Nava Álvarez, Maribel
- Abstract
Copyright of Congreso Internacional de Investigacion Academia Journals is the property of PDHTech, LLC and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2021
66. Modelo Híbrido de Minería de Datos para la Detección de Tendencias de Aprovechamiento Académico en Instituciones de Educación Superior.
- Author
-
Enrique Vivanco-Benavides, Luis, Fernando Bautista-López, Mauricio, Antonio Morales-Segundo, Luis, and Carlos Muñoz-Celaya, Roberto
- Abstract
Copyright of Congreso Internacional de Investigacion Academia Journals is the property of PDHTech, LLC and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2021
67. Diseño de un estimador basado en redes neuronales artificiales para caracterizar la frenada de un vehículo.
- Author
-
Garrosa, María, Olmeda, Ester, Sanz Sánchez, Susana, and Díaz, Vicente
- Subjects
ARTIFICIAL neural networks ,USER interfaces ,INFORMATION resources management ,PRESSURE sensors ,BRAKE systems ,ELECTRONIC data processing - Abstract
Copyright of Técnica Industrial: Revista Cuatrimestral de Ingeniería, Industria e Innovación is the property of Fundacion Tecnica Industrial and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2021
- Full Text
- View/download PDF
68. Predictive potential of the global bankruptcy models in the tourism industry.
- Author
-
del Castillo García, Agustín and Fernández Miguélez, Sergio Manuel
- Subjects
ARTIFICIAL neural networks ,TOURISM ,TRAVEL agents ,FINANCIAL crises ,HOTEL restaurants ,PREDICTION models ,BANKRUPTCY - Abstract
Copyright of Tourism & Management Studies is the property of Escola Superior de Gestao, Hotelaria e Turismo, Universidade do Algarve and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2021
- Full Text
- View/download PDF
69. El índice global de innovación en Colombia: un análisis y selección de los factores influyentes mediante el uso de redes neuronales artificiales.
- Author
-
Uribe Gómez, Julián Alberto
- Subjects
ARTIFICIAL neural networks ,FOREIGN investments ,HUMAN capital ,ENERGY consumption ,POSTSECONDARY education - Abstract
Copyright of Contaduría y Administración is the property of Facultad de Contaduria y Administracion-Universidad Nacional Autonoma de Mexico and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2021
- Full Text
- View/download PDF
70. Prediction of land degradation by Machine Learning Methods: A Case study from Sharifabad Watershed, Central Iran.
- Author
-
Habibi, Vahid, Ahmadi, Hassan, Jaffari, Mohammad, and Moeini, Abolfazl
- Subjects
- *
LAND degradation , *MACHINE learning , *WATER table , *ARTIFICIAL neural networks , *WATERSHEDS , *LEAST squares - Abstract
To monitor and predict the Groundwater levels in Sharifabad watershed, Central province, Iran three models of Partial Least Square Regression (PLSR), Artificial Neural Networks (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) have been used. In all models, 70% of the data was used for training, while 30% of data were employed for testing and validation. Monthly rainfall, topographic wetness index (TWI index), the distance from the river, Geographic location was the inputs and the level of groundwater was the output of each method. It is observed that ANN has the highest efficiency, which agrees with other findings. The results of ANN have been used in preparation of groundwater distribution map. According to the potential desertification map and groundwater level index, the potential of desertification had become severe since 2002 and was at a rate of 60% of land area, which, due to incorrect land management in 2016, increased to almost 98% of the land surface in the study area. Using ANN, it is predicted that 100% of the area was severely degraded for 2025. In addition to the target variable, latitude and longitude play important roles in ordinary Krigging and decreased the total error of two combined models. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF
71. Explainability in Models for Predicting the Probability of Default in Compliance with International Standard IFRS 9.
- Author
-
Murcia, Catalina Lozano and Romero, Francisco P.
- Abstract
Copyright of CISTI (Iberian Conference on Information Systems & Technologies / Conferência Ibérica de Sistemas e Tecnologias de Informação) Proceedings is the property of Conferencia Iberica de Sistemas Tecnologia de Informacao and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2021
72. RED NEURONAL ARTIFICIAL PARA EL CONTROL DE ROBOT DE SUPERVISIÓN INDUSTRIAL APLICADO EN LABORATORIO FARMACÉUTICO.
- Author
-
Bravo, Simón
- Abstract
The present work had as objective the design of an artificial neural network for the control of industrial supervision robot in pharmaceutical laboratory, which is supported by authors such as Aggarwal (2018), Berzal (2018), Krisel, D. (2007), Saeed, B. et al. (2020), among others. The methodology was descriptive projective with a non-experimental design. The population consisted of a Mobile Robot which presents the necessary variables to implement this control technique, from which a sample of distance was taken from its environment obtained by means of the ultrasonic sensors that this prototype has. Analysis matrixes were used as a data collection instrument. The results show that the controller was obtained using artificial neural networks necessary for the correct movement of the industrial supervision robot. Concluding that this control strategy can be applied with any model using a controller which has the necessary processing capacity for the application of the program and the modeling of the artificial neural network for the control of the automatic movement of the mobile robot. [ABSTRACT FROM AUTHOR]
- Published
- 2021
73. AVALIAÇÃO DE REDES NEURAIS PROFUNDAS PARA A PREVISÃO DE PREÇO DAS AÇÕES DA PETROBRÁS.
- Author
-
Lima de Campos, Lidio Mauro and Cardoso De Figueiredo, Yann Fabricio
- Abstract
Copyright of Revista Gestão & Tecnologia is the property of Revista Gestao & Tecnologia and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2021
74. Prediction of Compressive Strengths for Rice Husks Ash incorporated concrete, Using Neural Network and Reviews.
- Author
-
Ngunjiri Ngandu, Cornelius
- Published
- 2021
- Full Text
- View/download PDF
75. PREDICCIÓN DE AFINIDAD DE PÉPTIDOS CON LA MOLÉCULA MHC-II UTILIZANDO REDES NEURONALES ARTIFICIALES.
- Author
-
Merecías Pérez, Jacob Ezequiel, Castro Liera, Marco Antonio, Angulo, Carlos, and Morales Viscaya, Joel Artemio
- Abstract
Copyright of Congreso Internacional de Investigacion Academia Journals is the property of PDHTech, LLC and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2020
76. Redes Neuronales Artificiales Aplicadas al Análisis de Datos de Pymes Utilizando Tecnología Blockchain.
- Author
-
Acosta Mendizábal, Marco Antonio, Andrade Cruz, Mario Jesús, Quintero Aguilar, Verónica Belem, Pérez Sánchez, Johany Yuliet, Solares Sánchez, Abril, and Nava Álvarez, Maribel
- Abstract
Copyright of Congreso Internacional de Investigacion Academia Journals is the property of PDHTech, LLC and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2020
77. ANÁLISIS DEL COMPORTAMIENTO DE TEMPERATURA DE UN MOTOR DE COMBUSTIÓN INTERNA USANDO REDES NEURONALES ARTIFICIALES.
- Author
-
Trujillo Jiménez, Piña Castillo, Juan José, Jiménez Macedo, Víctor Daniel, and González Bernal, Renato
- Abstract
Copyright of Congreso Internacional de Investigacion Academia Journals is the property of PDHTech, LLC and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2019
78. Predicción de precipitación mensual mediante Redes Neuronales Artificiales para la cuenca del río Cali, Colombia
- Author
-
Daniel David Montenegro Murillo, Mayra Alejandra Pérez Ortiz, and Viviana Vargas Franco
- Subjects
reducción de escala ,redes neuronales artificiales ,escenarios de cambio climático ,Technology ,Mining engineering. Metallurgy ,TN1-997 - Abstract
Estudiar el comportamiento futuro de la precipitación en las cuencas hidrográficas es un tema vital cuando se habla realizar un correcto ordenamiento territorial de las mismas, ya que, esto permitiría disminuir la vulnerabilidad y mitigar desastres. Por esta razón, este estudio se enfocó realizar un análisis de los escenarios de cambio climático en la cuenca hidrográfica del río Cali; partiendo de una base datos de precipitación mensual de 35 estaciones y Modelos de Circulación General (GCM) del conjunto de datos CMIP5, a partir de estos se realizó una reducción de escala estadística de los escenarios RCP 2.6, 4.5 y 8.5 mediante Redes Neuronales Artificiales y posteriormente se analizaron los cambios que se presentaran para el año 2100. Estos análisis permitieron establecer que los diferentes escenarios analizados afirman que en los años venideros existirá un desplazamiento de la precipitación de la zona alta a la media y baja de la cuenca.
- Published
- 2019
- Full Text
- View/download PDF
79. Diseño de un sistema de reconocimiento de patrones en imágenes termográficas y de huella plantar para la identificación de pie plano en niños con edades entre cinco y seis años
- Author
-
Milton Javier Muñoz-Neira, Anyed Stephany Martínez-Parra, Cristian Gerardo Ruiz-Adarme, Carlos Humberto Triana-Castro, and Jorge Luis Cornejo-Plata
- Subjects
pie plano ,patrones de textura ,patrones de huella ,redes neuronales artificiales ,Technology ,Science - Abstract
The following paper presents the main results of exploratory research-oriented to the design and implementation of a pattern recognition system for flatfoot identification in children between 5 and 6 years. Patterns were determined from texture analysis of foot thermographic images, and from contour analysis of footprint images. For each case, an artificial neuronal network was trained, with base in a back-propagation algorithm. In each trial, 70 % of data were used for training, and 30 % for validation. For experiments done, success rates greater than 80 % were achieved. The best results were reached with contour patterns reduced by principal components analysis, PCA, in a binary system, with a success rate of 90.84 % in cross-validation. Results are a contribution to the study of diagnostic techniques for flatfoot treatment through the use of technologic tools.
- Published
- 2019
- Full Text
- View/download PDF
80. Artificial Neural Model based on radial basis function networks used for prediction of compressive strength of fiber-reinforced concrete mixes
- Author
-
Luis Octavio González Salcedo, Aydee Patricia Guerrero Zúñiga, Silvio Delvasto Arjona, and Adrián Luis Ernesto Will
- Subjects
hormigón reforzado con fibras ,resistencia a la compresión ,predicción de propiedades ,redes neuronales artificiales ,funciones de base radial ,inteligencia artificial ,Engineering (General). Civil engineering (General) ,TA1-2040 ,Science - Abstract
Existe una relación compleja y no lineal entre los factores que influyen en la resistencia de diseño y la compresión de hormigones reforzados con fibras de acero. La relación entre las variables de entrada, los factores y la variable de salida, y la resistencia de diseño a la compresión puede ser obtenida por un modelo neuronal artificial, cuyas características sean autoadaptación, autoestudio y mapeo no lineal. En este documento se presenta la elaboración de un modelo neuronal artificial basado en redes neuronales de funciones de base radial. La resistencia de diseño a la compresión en dosificaciones de mezclas de hormigón reforzados con fibras de acero es estimada, predicción que se analiza a partir del coeficiente de correlación R al compararse con los valores reales de la resistencia. Los resultados muestran que los valores estimados usando las redes de base radial coinciden con los valores experimentales, y la capacidad de predicción de la propiedad mecánica del modelo neuronal es mejor que la de otros modelos basados en redes multicapas desarrollados por los autores. El entrenamiento de los modelos neuronales permitió concluir que el uso de relaciones de los materiales es un indicador más adecuado para la comparación entre diferentes dosificaciones de mezclas de hormigón que llevan a similares resistencias a la compresión. Así, se potencia una agenda futura en la generación de nuevos métodos de estudio de la resistencia de diseño a la compresión en hormigones reforzados con fibras metálicas en el campo de la ingeniería.
- Published
- 2019
- Full Text
- View/download PDF
81. Estudio de redes neuronales para el pronóstico de la demanda de asignaturas
- Author
-
Jesús David Terán-Villanueva, Salvador Ibarra-Martínez, Ph. D., Julio Laria-Menchaca, Ph. D., José Antonio Castán-Rocha, Ph. D., Mayra Guadalupe Treviño-Berrones, M.Sc., Alejandro Humberto García-Ruiz, and José Eduardo Martínez-Infante
- Subjects
planeación estratégica ,pronóstico de demanda ,redes neuronales artificiales ,Engineering (General). Civil engineering (General) ,TA1-2040 - Abstract
La planeación de cursos de un centro educativo o universidad está compuesta por múltiples problemas complejos como lo es la asignación de horarios para los alumnos, salones y profesores para cada asignatura. Uno de los problemas iniciales es determinar la cantidad de asignaturas que se ofertarán; este problema parece sencillo a simple vista ya que una vez que se tenga la información de la cantidad de alumnos aprobados para cada asignatura, se puede calcular fácilmente la siguiente demanda de asignaturas. Sin embargo, existen ocasiones en los que la planeación de cursos del siguiente período inicia antes de tener la información relativa a la aprobación de los alumnos. Lo cual nos lleva al problema del pronóstico de los porcentajes de aprobación para calcular la demanda de asignaturas de los alumnos. En este trabajo se compara el desempeño de modelos causales contra modelos estadísticos para el pronóstico de los porcentajes de aprobación y reprobación de los alumnos. Los resultados finales muestran una ventaja importante de los métodos causales sobre los métodos estadísticos para los casos de prueba. Consideramos que esta ventaja ocurre debido a que el modelo causal aprende los patrones de comportamiento de los datos de entrenamiento de forma independiente en vez de generalizar porcentajes de acreditación. Además de lo anterior, el método estadístico puede presentar problemas importantes al tratar de pronosticar porcentajes de acreditación para situaciones que no se encuentren en los datos de entrenamiento, mientras que el modelo causal utilizará la información aprendida para pronosticar dichas situaciones.
- Published
- 2019
- Full Text
- View/download PDF
82. Artificial intelligence techniques for optimizing the efficiency in public works contracting selection processes
- Author
-
Reiner Solís-Villanueva
- Subjects
redes neuronales artificiales ,toma de decisiones ,riesgo ,Systems engineering ,TA168 - Abstract
The present article proposes to design a model that provides a generic architecture which acts autonomously in public works contracting selection processes, in order to generate an automated decision criterion in the event of a tie. For the Simplified Tender selection process, in case of a tie, it is proposed to choose the bidder by means of an electronic lottery based on a controlled randomization system of encryption and transformation. For the Public Bidding selection process, in the event of a tie, the bidder is chosen by means of a predicted compliance index according to the behavior of the companies when executing similar infrastructure projects. To this end, a model that predicts the probability of success or failure of the bidder to execute a project before initiating it is generated, using artificial neural networks as an analysis tool. This paper reviews the common characteristics of artificial neural networks.
- Published
- 2018
- Full Text
- View/download PDF
83. Predicción por redes neuronales artificiales del peso corporal de Capra hircus en crianza semiextensiva
- Author
-
Miguel Callacná-Custodio, Joan Díaz-Huamanchumo, and Victor Vásquez-Villalobos
- Subjects
Redes neuronales artificiales ,caprinos criollos mejorados ,sistema de crianza ,peso corporal ,regresión múltiple ,Agriculture (General) ,S1-972 ,Technology ,Industrial engineering. Management engineering ,T55.4-60.8 - Abstract
Objetivo del presente trabajo fue predecir por redes neuronales artificiales (RNA) el peso corporal de caprinos en crianza semiextensiva. Se utilizó 40 caprinos criollos mejorados desde el nacimiento hasta las seis semanas de edad. El 80% de la data fue utilizada para entrenar la red y el 20 % para validarla. El tipo de RNA usada fue del tipo feedforward (FF), con algoritmo de entrenamiento Backpropagation (BP) y ajuste de pesos Levenberg–Marquardt (LM), topología que presento el mejor resultado: 3 entradas, seis salidas lineales (purelin), capa oculta con 42 neuronas, tasa de aprendizaje de 0,01, coeficiente de momento de 0,5, meta del error de 0,0001 y 100 etapas de entrenamiento. Comparativamente el error porcentual promedio de los valores predichos por la RNA fue de 7,51 y por la regresión múltiple de 7,80 no existiendo diferencia significativa entre ambos (p > 0,05). Así mismo, el porcentaje de aciertos de la RNA fue de 50% y de la regresión múltiple de 50%, mostrando en ambos casos un rendimiento similar.
- Published
- 2018
- Full Text
- View/download PDF
84. Impacto de la estandarización y escalado: factor para predicción de costos en proyectos a través de una red neuronal artificial.
- Author
-
Rodríguez González, Joselyn and Ugalde Saborio, Edgar
- Subjects
- *
ARTIFICIAL neural networks , *ELECTRONIC data processing , *PROJECT managers , *MACHINE learning , *STANDARDIZATION - Abstract
This paper presents comparison of Standardization and Scaling methods in cost predicting. Four methods were used for pre-processing dataset; after that, data was processed through artificial neural network (RNA). The first step was to build common variables in information projects based on opinions of some project managers. Second step was to simulate dataset based on information provided by CRConsulting. Third one was process data with machine learning according to the four methods proposed, RNA algorithms were the same in four cases. Last, the comparison results were presented through adjustment models according to the applied method. [ABSTRACT FROM AUTHOR]
- Published
- 2021
85. Distribución cortical de la potencia absoluta de la actividad Beta 12Hz-25 Hz en niños varones con trastorno por déficit de atención e hiperactividad combinado.
- Author
-
Téllez-Villagra, C. and Pedraza Avilés, A. González
- Abstract
Introduction: The Beta rhythm of the quantitative electroencephalogram (QEEG) is associated with inattention and movement disorders. In children with attention déficit hyperactivity disorder (ADHD), absolute-power (AP) with an increase in slow frequencies and a decrease in fast, especially Beta-total, has been reported. Objective: To identify the cortical distribution of PA decreased or increased in the QEEG at rest eyes closed of each Beta frequency (12Hz-25Hz) as predictor of visual or auditory and initiation or inhibition of movement inattention, in boys with combined ADHD. Material and Methods: Retrospective study (2008-2019) examining the medical records of 131 boys, 6-14 years, with a diagnosis of combined ADHD. For each, child 532 data were obtained: PA + 2 of the norm (Neuroguide base), Beta (12-25Hz) in 19 QEEG derivations were associated with visual, auditory and movement inattention (score <80 in TOVA-Visual-and-Auditory). Results: AP decreased in 1738 referrals (81.5%); AP-increased in 394 (18.48%), Beta 20-25Hz AP decreased predominated in Frontal and Central-temporaloccipital; 12-13Hz AP increased in Parietal. Visual inattention was lower than auditory. Visual Variability and Response Time characterized the poor performance. AP decreased Beta 25Hz in Frontal characterized 30 (43%) children with visual and auditory inattention; Beta 23-25Hz in Center-temporal-occipital characterized 33 (75%) individuals with visual inattention; AP increased 21Hz in Frontal and 25Hz in Parietal characterized 2 (29%) children with auditory inatention. Beta 13-25Hz AP-decreased in Frontal and Centraltemporal-occipital and 20-25Hz in Parietal inf luenced visual inattention in all its variables; while auditory inattention in all its variables was inf luenced by Beta 16-25Hz in Center-temporaloccipital. Beta 16-25Hz AP decreased in Frontal and Central-temporal-occipital inf luenced visual and auditory hyperactivity; Beta 22-25Hz in Centraltemporal-occipital inf luenced visual and auditory impulsivity. Conclusion: Beta 20-25Hz with AP decreased in Central-temporal-occipital and 12-13Hz with AP increased in Parietal, together with Visual Variability and Response Time, could be biomarkers of combined ADHD. The biomarkers could support a more precise diagnosis and the use of non-pharmacological therapy with state-of-the-art technology focused on regulating brain electrical activity. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF
86. Modeling Cutting Forces in High-Speed Turning using Artificial Neural Networks.
- Author
-
Hernández-González, Luis W., Curra-Sosa, Dagnier A., Pérez-Rodríguez, Roberto, and Zambrano-Robledo, Patricia D. C.
- Published
- 2021
- Full Text
- View/download PDF
87. Análisis comparativo de susceptibilidad de erosión y evaluación de incertidumbre en la subcuenca del Río Claro, Costa Rica
- Author
-
Iván Pérez-Rubio, Daniel Flores, Christian Vargas, Andreas Mende, and Francisco Jiménez
- Subjects
Ganadería ,redes neuronales artificiales ,regresión logística ,subcuenca hidrográfica ,teledetección ,Human ecology. Anthropogeography ,GF1-900 ,Natural history (General) ,QH1-278.5 - Abstract
[Introducción]: La deforestación y la gestión insostenible de los sistemas de producción agrícola y ganadero en áreas montañosas han provocado la degradación de la tierra y una progresiva reducción en la provisión de los servicios ecosistémicos. [Objetivo]: En este artículo se desarrolla un análisis espacial de susceptibilidad de erosión en la subcuenca del río Claro, en el cordón montañoso Fila Cruces, en la región del Pacífico Sur de Costa Rica. [Metodología]: Para ello se aplicaron los métodos de regresión logística y redes neuronales artificiales integrados en un entorno de sistemas de información geográfica (GIS) y empleando herramientas de teledetección. En ambos modelos se consideraron los siguientes factores explicativos: uso del suelo, geomorfología, pendiente, distancia euclidiana a la red de drenaje e índice de vegetación diferencial normalizado (NDVI, en sus siglas en inglés). Los mapas de susceptibilidad de erosión fueron validados independientemente por medio de la función características operativas del receptor (ROC, por sus siglas en inglés). [Resultados]: El modelo de redes neuronales artificiales obtuvo un poder predictivo superior al de regresión logística con base en el valor calculado del área debajo de la curva (AUC, por sus siglas en inglés). Los factores con mayor poder explicativo variaron en función del modelo utilizado [Conclusiones]: Los mapas de susceptibilidad de erosión mostraron una elevada alteración ecológica en términos de la probabilidad de ocurrencia de procesos de erosión, especialmente en la parte alta de la subcuenca, en terrenos ocupados por fincas de ganadería extensiva y elevada pendiente.
- Published
- 2021
- Full Text
- View/download PDF
88. Redes neuronales artificiales aplicadas en sistemas de predicción para la seguridad vial
- Author
-
Karina Patricia Carpio and Fernando Oñate Valdivieso
- Subjects
redes neuronales artificiales ,accidentes de tránsito ,predicción ,elementos viales ,herramienta ,red neuronal multicapa. ,Technology ,Science - Abstract
En esta investigación se aplican redes neuronales artificiales para el análisis de variables que podrían tener influencia en la ocurrencia de accidentes de tránsito en carreteras de montaña. El modelo se desarrolló utilizando Redes Neuronales Artificiales (RNA) y se evaluó su eficiencia para la predicción de accidentes de tránsito considerando variables como radios de curvatura y pendientes de la vía. El desempeño de las Redes Neuronales Artificiales se evaluó aplicando la eficiencia de Nash- Sutcliffe y en el error cuadrático medio. Los resultados de la investigación mostraron un bajo desempeño de las Redes Neuronales Artificiales en el pronóstico de accidentes en función de las variables seleccionadas, lo que sugiere que los accidentes aparentemente no se deben a la geometría de la vía, o topografía, geografía del terreno, sino a otros elementos tales como el exceso de velocidad o la impericia del conductor. Sin embargo, la investigación muestra varias alternativas de modelamientos de la Red para intentar tener una mejor predicción.
- Published
- 2020
- Full Text
- View/download PDF
89. Machine learning for carbon stock prediction in a tropical forest in Southeastern Brazil.
- Author
-
Dantas, Daniel, Nunes Santos Terra, Marcela de Castro, Baldissera Schorr, Luis Paulo, and Calegario, Natalino
- Subjects
- *
TROPICAL forests , *CARBON sequestration in forests , *MACHINE learning , *ARTIFICIAL neural networks , *SUPPORT vector machines - Abstract
The increasing awareness of global climate change has drawn attention to the role of forests as mitigators of this process as they act as carbon sinks to the atmosphere. Understanding the process of carbon storage in forests and its drivers, as well as presenting consistent models for their estimation, is a current demand. In this sense, the aim of this study was to evaluate the performance of machine learning techniques: support vector machines (SVM) and to propose a new nonlinear model extracted from the training of an artificial neural network (ANN) in the modeling of above ground carbon stock in a secondary semideciduous forest. SVM and ANN construction and training process considered independent variables selected by stepwise: minimum DBH (diameter of breast height - 1.3 m), maximum DBH, mean DBH, total height and number of trees, all by plot. SVM and the model extracted from ANN were applied to the data set intended for validation. Both techniques presented satisfactory performance in modeling carbon stock by plot, with homogeneous distribution and low dispersion of residues and predicted values close to those observed. Analysis criteria indicated superior performance of the model extracted from the artificial neural network, which presented a mean relative error of 6.94 %, while the support vector machine presented 13.52 %, combined with lower bias values and higher correlation between predictions and observations. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF
90. Treatment of individual predictors with neural network algorithms improves Global Registry of Acute Coronary Events score discrimination.
- Author
-
Borracci, Raul A., Higa, Claudio C., Ciambrone, Graciana, and Gambarte, Jimena
- Subjects
- *
NEURAL circuitry , *DISCRIMINATION (Sociology) , *ACUTE coronary syndrome , *LOGISTIC regression analysis , *CREATINE kinase - Abstract
Objective: The aim of this study was to develop, train, and test different neural network (NN) algorithm-based models to improve the Global Registry of Acute Coronary Events (GRACE) score performance to predict in-hospital mortality after an acute coronary syndrome. Methods: We analyzed a prospective database, including 40 admission variables of 1255 patients admitted with the acute coronary syndrome in a community hospital. Individual predictors included in GRACE score were used to train and test three NN algorithm-based models (guided models), namely: one- and two-hidden layer multilayer perceptron and a radial basis function network. Three extra NNs were built using the 40 admission variables of the entire database (unguided models). Expected mortality according to GRACE score was calculated using the logistic regression equation. Results: In terms of receiver operating characteristic area and negative predictive value (NPV), almost all NN algorithms outperformed logistic regression. Only radial basis function models obtained a better accuracy level based on NPV improvement, at the expense of positive predictive value (PPV) reduction. The independent normalized importance of variables for the best unguided NN was: creatinine 100%, Killip class 61%, ejection fraction 52%, age 44%, maximum creatine-kinase level 41%, glycemia 40%, left bundle branch block 35%, and weight 33%, among the top 8 predictors. Conclusions: Treatment of individual predictors of GRACE score with NN algorithms improved accuracy and discrimination power in all models with respect to the traditional logistic regression approach; nevertheless, PPV was only marginally enhanced. Unguided variable selection would be able to achieve better results in PPV terms. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF
91. Classification of the Angular Position During Wrist Flexion-Extension Based on EMG Signals.
- Author
-
Fajardo-Perdomo, María Alexandra, Guardo-Gómez, Verónica, Orjuela-Cañon, Alvaro David, and Ruiz-Olaya, Andrés Felipe
- Subjects
- *
WRIST , *JOB classification , *ARTIFICIAL neural networks , *PATTERN recognition systems , *AUTOREGRESSIVE models , *REHABILITATION technology - Abstract
Objective: To evaluate a group of features in a myoelectric pattern recognition algorithm to differentiate between five angular positions of the wrist during flexion-extension movements. Materials and Methods: An experimental configuration was made to capture the EMG and wrist joint angle related to flexion-extension movements. After that, a myoelectric pattern recognition algorithm based on a multilayer perceptron artificial neural network (ANN) was implemented. Three different groups were used: Time domain characteristics, autoregressive (AR) model parameters, and representation of time frequency using Wavelet transform (WT). Results and Discussion: The experimental results of 10 healthy subjects indicate that the coefficients of the AR models offer the best parameters for classification, with a differentiation rate of 78 % for the five angular positions studied. The combination of frequency and time frequency resulted in a differentiation rate that reached 82 %. Conclusions: An algorithm based on pattern recognition of EMG signals was used to carry out a comparative study of groups of features that allow for the differentiation of the angular position of the wrist in terms of flexion-extension movements. The method has the potential for application in the field of rehabilitation engineering to detect the user's movement intent. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF
92. Selección organizacional: resiliencia y desempeño de las pymes en la era de la COVID-19.
- Author
-
García-Contreras, Rigoberto, Valle-Cruz, David, and Azalea Canales-García, Rosa
- Subjects
COVID-19 pandemic ,ORGANIZATIONAL resilience ,ARTIFICIAL neural networks ,ORGANIZATIONAL performance ,DATA analysis - Abstract
Copyright of Estudios Gerenciales is the property of Universidad ICESI and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2021
- Full Text
- View/download PDF
93. Doble evaluación de la susceptibilidad por movimientos en masa basada en redes neuronales artificiales y pesos de evidencia.
- Author
-
Goyes-Peñafiel, Paul and Hernandez-Rojas, Alejandra
- Subjects
ARTIFICIAL neural networks ,PRINCIPAL components analysis ,LOGISTIC regression analysis ,HAZARD mitigation ,QUANTITATIVE research ,LANDSLIDES ,RECEIVER operating characteristic curves - Abstract
Copyright of Boletin de Geologia is the property of Universidad Industrial de Santander and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2021
- Full Text
- View/download PDF
94. Forest-Genetic method to optimize parameter design of multiresponse experiment.
- Author
-
Villa-Murillo, Adriana, Carrión, Andrés, and Sozzi, Antonio
- Subjects
- *
EXPERIMENTAL design , *GENETIC algorithms , *SIMULATED annealing - Abstract
We propose a methodology for the improvement of the parameter design that consists of the combination of Random Forest (RF) with Genetic Algorithms (GA) in 3 phases: normalization, modelling and optimization. The first phase corresponds to the previous preparation of the data set by using normalization functions. In the second phase, we designed a modelling scheme adjusted to multiple quality characteristics and we have called it Multivariate Random Forest (MRF) for the determination of the objective function. Finally, in the third phase, we obtained the optimal combination of parameter levels with the integration of properties of our modelling scheme and desirability functions in the establishment of the corresponding GA. Two illustrative cases allow us to compare and validate the virtues of our methodology versus other proposals involving Artificial Neural Networks (ANN) and Simulated Annealing (SA). [ABSTRACT FROM AUTHOR]
- Published
- 2020
- Full Text
- View/download PDF
95. Bearing Capacity and Settlement Prediction of Multi-Edge Skirted Footings Resting on Sand.
- Author
-
Gnananandarao, Tammineni, Khatri, Vishwas Nandkishor, and Dutta, Rakesh Kumar
- Published
- 2020
- Full Text
- View/download PDF
96. Trombosis venosa profunda en extremidades inferiores: revisión de las técnicas de diagnóstico actuales y su simbiosis con el aprendizaje automático para un diagnóstico oportuno
- Author
-
Maria Berenice Fong-Mata, Everardo Inzunza-González, Enrique Efrén García-Guerrero, David Abdel Mejía Medina, Oscar Adrián Morales Contreras, and Antonio Gómez-Roa
- Subjects
Diagnóstico ,Redes neuronales artificiales ,Trombosis venosa profunda. ,Engineering (General). Civil engineering (General) ,TA1-2040 ,Technology (General) ,T1-995 - Abstract
La Trombosis Venosa Profunda (TVP) es una manifestación de una Enfermedad Tromboembólica (ET). Cuando en una TVP los trombos venosos se desprenden y viajan a través del torrente sanguíneo pueden ocasionar una Trombo Embolia Pulmonar (TEP). La existencia de Trombosis Venosa Profunda (TVP) en las extremidades inferiores se ha descrito como uno de los principales factores de riesgo para el desarrollo de la TEP. Se considera que hasta el 90% de los émbolos pulmonares proceden de trombos venosos de las extremidades inferiores. Las técnicas más utilizadas para la detección de TVP son los modelos de probabilidad clínica, el dímero D y las pruebas de imagen no invasivas, como la ecografía para la TVP y la angiotomografía computadorizada (TC) para el embolismo pulmonar. Sin embargo, debido a la inespecificidad de los síntomas de la TVP, el umbral para ordenar una ecografía es bajo, además de ser un proceso complicado que requiere la participación de un médico especialista para su interpretación. En las últimas décadas el aprendizaje automático ha surgido como apoyo en la toma de decisiones para el diagnóstico de diversas enfermedades, algunas de las tecnologías más utilizadas en el campo de la medicina incluyen Support Vector Machine (SVM), Árboles de decisión y las Redes Neuronales Artificiales (RNA). En el presente artículo se hace una revisión de las tecnologías existentes para la detección de la TVP así como de los principales algoritmos de aprendizaje automático comúnmente utilizados en aplicaciones biomédicas; se propone el diseño de un sistema computarizado que utilice técnicas de aprendizaje automático como herramienta de apoyo para la detección oportuna de un posible padecimiento de TVP.
- Published
- 2020
- Full Text
- View/download PDF
97. Estudio experimental y simulación del comportamiento inelástico de paneles compuestos usando redes neuronales artificiales
- Author
-
Wilmer Barreto and Ricardo Picón
- Subjects
daños ,redes neuronales artificiales ,análisis no lineal ,paneles compuestos ,estructuras no tradicionales ,desplazamientos permanentes ,experimentos a flexión ,Architecture ,NA1-9428 ,Building construction ,TH1-9745 - Abstract
El análisis de estructuras complejas, como los paneles compuestos de varios materiales, es difícil de modelar producto de la variabilidad en las propiedades mecánicas de los materiales. Lo anterior, aunado a la no-linealidad en el comportamiento de los materiales, hace que la aplicación de los métodos tradicionales de cálculo numérico sea difícil y demande mayor tiempo de cómputo. El presente trabajo introduce técnicas menos convencionales de cómputo como lo son las redes neuronales artificiales (RNA) para la modelación de la deformación permanente y daños en una losa compuesta sujeta a flexión. Se entrenaron y verificaron 400 modelos de RNA, los cuales fueron capaces de modelar la no linealidad del elemento estructural a flexión, reproducir exitosamente los daños por agrietamiento y pandeo del panel, así como reproducir la deformación permanente global del elemento estudiado.
- Published
- 2020
- Full Text
- View/download PDF
98. Corrección del pronóstico cuantitativo de la precipitación mediante el uso de redes neuronales
- Author
-
A. Fuentes, M. Sierra, and Y. Morfa
- Subjects
redes neuronales artificiales ,pronóstico cuantitativo de precipitación ,WRF ,corrección de sesgos ,Meteorology. Climatology ,QC851-999 - Abstract
En el presente trabajo se propone un modelo de redes neuronales como una técnica eficaz para la corrección del pronóstico cuantitativo de precipitación brindado por el modelo WRF. Para ello se emplea un Perceptrón Multi-Capa, con el objetivo de utilizar la salida (observaciones brindadas por las estaciones en superficie) para establecer una relación con los elementos de entrada (salidas del WRF). Se realiza el entrenamiento del modelo con datos reales de acumulado de precipitación correspondientes al año 2017; y se realiza la evaluación con el período comprendido entre el 4 de noviembre de 2018 y el 28 de febrero de 2019. Se logró la corrección del pronóstico cuantitativo de la precipitación en las estaciones analizadas, siendo más significativa la mejoría para la estación de Montaña y en los casos en los que el WRF sobreestima el acumulado de precipitación.
- Published
- 2020
99. Modelización y predicción espacio-tiempo de la irradiancia solar global a corto plazo mediante redes neuronales artificiales y geoestadística
- Author
-
Federico Vladimir Gutiérrez-Corea, Miguel Ángel Manso-Callejo, and Francisco Serradilla-García
- Subjects
Redes neuronales artificiales ,geoestadística ,pronóstico. ,Maps ,G3180-9980 ,Cartography ,GA101-1776 - Abstract
Mejorar el conocimiento de la Irradiancia Solar (IS) sobre la superficie terrestre, así como su predicción (pronóstico), cobra especial interés por su importancia para las energías renovables, a como lo son los sistemas basados en Energía Solar (ES), y para distintas aplicaciones industriales o ecológicas. En la presente investigación se ha experimentado con cinco técnicas de estimación espacial de la IS a intervalos de 15 minutos, en el territorio peninsular español, con distintas configuraciones espaciales. Encontrándose que la geoestadística mediante el Kriging con Regresión, usando variables auxiliares -una de ellas la IS estimada a partir de imágenes satelitales- permite estimar espacialmente la IS más allá de los 25 km, identificados en las investigación científicas previas, como límite de distancia máxima al punto de estimación. Se ha experimentado con el modelado de Redes Neuronales Artificiales (RNA) para la predicción en tiempo futuro (temporal) -a corto plazo- de la IS utilizando observaciones próximas (componentes espaciales) en sus entradas y los resultados son prometedores. Así los niveles de errores disminuyen, en relación a investigaciones relacionadas, bajo las siguientes condiciones: cuando el horizonte temporal de predicción es inferior o igual a 3 horas, las estaciones vecinas que se incluyen en los modelos deben encentrarse a una distancia máxima aproximada de 55 km.
- Published
- 2020
- Full Text
- View/download PDF
100. Estudio comparativo de los algoritmos backpropagation (bp) y multiple linear regression (mlr) a través del análisis estadístico de datos aplicado a redes neuronales artificiales
- Author
-
Iván Mesias Hidalgo-Cajo, Saul Yasaca-Pucuna, Byron Geovanny Hidalgo-Cajo, Diego Patricio Hidalgo-Cajo, and Nelly Baltazara Latorre-Benalcázar
- Subjects
Backpropagation ,Multiple Linear Regression ,Redes neuronales artificiales ,Education (General) ,L7-991 - Abstract
El objetivo de la investigación es comparar el algoritmo Backpropagation desarrollado por el usuario bajo software libre Java y el algoritmo Multiple Linear Regression, dicha comparación demanda del análisis estadístico decriptivo basado en redes neuronales artificiales. Se utilizó específicamente dos modelos de algoritmos de predicción aplicados a 451 patrones o registros a procesar que están repartidos en las primeras 401 filas para entrenamiento de la red neuronal y los otros 50 registros para validación y prueba, conformado por 4 variables de entrada (Height above sea level, Fall, Net fall, Flux) y 1 variable a predecir (Power turbine), para las diferentes pruebas los parámetros de entrenamiento y selección con los mejores resultados son: Architecture of the neural network, Type of scaling of data, Initial range of weight and thresholds, Learning rate and Momentum, Batched / online, Number of training epochs. Entre los resultados de comparación de los algoritmos analizados se determinó que el error en mayor iteracciones es menor que son respuestas de los 50 patrones de prueba. En el algoritmo Multiple Linear Regression la variable real es el valor de la variable a predecir, esta variable es la suministrada a predecir por el usuario y es el valor que se predijo de la red neuronal, la variable prediction es la diferencia que se hace de la resta de los anteriores errores y se lo realiza para calcular el error y el total error es el valor mínimo a obtener que representa el error calculado de todos los datos, es decir el porcentaje de error de la red neuronal de back-propagation. Entre más bajo es este porcentaje mejor será la red, porque menor será su porcentaje de error.
- Published
- 2020
- Full Text
- View/download PDF
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.