Search

Alzheimer's disease is a neurological disorder that causes loss of autonomy and memory in people who suffer from it. Due to the increase in cases of this disease and the lack of accuracy of diagnostic tools, new tools capable of addressing this issue are being developed. The main objective of this research work is to implement a three-dimensional convolutional neural network model with AlexNet3D type base structure to predict the possible diagnosis of Alzheimer's disease (AD). from the analysis of magnetic resonance images, using as an early stage mild cognitive impairment syndrome (MCI). The construction phases of the project will be explained, divided into database selection, feature selection, data processing, model development for training and validation, and finally, results obtained from prediction tests. With this it was possible to obtain a percentage of 72,222 %, allowing the K-Net95 model to be cataloged as a stable and efficient network, despite the computational limitations to which the project was limited. [ABSTRACT FROM AUTHOR]

Copyright of Revista Entramado is the property of Universidad Libre Seccional Cali 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.)

Copyright of DYNA - Ingeniería e Industria is the property of Publicaciones Dyna SL 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.)

Copyright of DYNA - Ingeniería e Industria is the property of Publicaciones Dyna SL 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.)

In this work, a review of the existing paradigms and the most used techniques in the burst detection is carried out, delving into those that use Machine Learning as the main tool for data interpretation. The relationship between detection effectiveness and the parameters of each algorithm, as well as the level of processing required, are compared. For Support Vector Machine, the effectiveness in burst detection is exponentially related to the number of combinations of C and γ. The exposed decision tree increases its precision the more information about the state of the network it has. The artificial neural network demonstrates a detection effectiveness at the level of the rest of the algorithms treated, maintaining the commitment to the processing level. [ABSTRACT FROM AUTHOR]

University spaces generate entrepreneurial ideas that manifest themselves in classroom tasks, and some of them materialize as real ventures. The aim of this research is to understand how the creative characteristics present in the academic environment impact Technological Readiness Level (TRL) development. It is a correlational cross-sectional study with a quantitative, non-experimental approach, and a non-probabilistic consecutive sample of 384 subjects obtained through an instrument validated by expert judgment, covering 77% of Higher Education Institutions in the studied territory. Through exploratory factor analysis with SPSS and confirmatory analysis with AMOS, speci- fic knowledge, motivation for entrepreneurship, and sought impact are identified as determining creative factors; gender identity shows no incidence. Additionally, using Artificial Neural Networks (ANN), a model proposing the normalized importance of factors with 67.2% assertiveness is suggested. [ABSTRACT FROM AUTHOR]

Copyright of Revista Argentina de Cardiología is the property of Sociedad Argentina de Cardiologia 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.)

The Mesa de Los Santos region is known for its continuous development in infrastructure, agriculture, and tourism, especially in the north sector, which has generated a greater demand in the supply of drinking water. Previous geological, hydrological, and geophysical studies highlighted the importance of understanding the physical properties of rocks, as well as their spatial estimation, the reason why we implemented neural network models in this work. The laboratory results for the samples associated with the Los Santos Formation present porosities of up to 25%, %, while the Rosa Blanca Formation shows values of less than 5% in carbonate units and up to 30% for sandy lithologies. The P-wave velocity for both formations of interest is in a range between 1000 y 6000 m · s-1. The resistivity values were influenced by the pore fluid (water), obtaining values between 10² - 104 Ω · m for the Los Santos and Rosa Blanca formations. The maps generated by the neural network provided a consistent and accurate geological distribution compared to traditional interpolation methods. However, uncertainty was noted in areas with intense fracturing and weathering, which could affect the results in areas close to geological faults. These findings provide valuable information for geological, hydrogeological, and engineering studies in the region. [ABSTRACT FROM AUTHOR]

The Center of Molecular Immunology is an exponent of Cuban biotechnology institution dedicated to basic research, development, production and commercialization of biopharmaceuticals, with the aim of diagnosing and treating cancer and diseases related to the immune system. In this work, a bibliographic synthesis of research that was carried out in this center in the period from 2012 to 2018, was performed, with the purpose of analyzing the application of some Data Mining techniques in the evaluation of fermentation and purification stages of the process for obtaining biopharmaceuticals in three production facilities. The phases of Data Mining were characterized and the practical implication of the results was presented. Principal Components Analysis was applied as a descriptive model using THE UNSCRAMBLER software, and Artificial Neural Networks were applied as a predictive model using the MATLAB neural networks toolbox. The use of these models made it possible to extract useful information for decision making, thereby explaining the behavior of the parameters that influence the quality of the final product, as well as estimating important variables such as the concentration of the protein of interest in the fermentation supernatant and the performance of the purification stage, depending on the process variables that have the greatest influence on its behavior. [ABSTRACT FROM AUTHOR]

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.)

Copyright of Avances en Ciencias e Ingeniería is the property of Avances en Ciencias e Ingenieria 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.)

Copyright of Revista Iberoamericana de Ingeniería Mecánica is the property of Editorial UNED 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.)

Copyright of Revista ION 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.)

Copyright of Religación: Revista de Ciencias Sociales y Humanidades is the property of Religacion: Revista de Ciencias Sociales y Humanidades 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.)

Copyright of Revista Cubana de Ciencias Informáticas is the property of Universidad de las Ciencias Informaticas (UCI) 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.)

Copyright of Revista Sociedad Colombiana de Oftalmología is the property of Sociedad Colombiana de Oftalmologia 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.)

Copyright of Acta Comportamentalia is the property of Instituto de Psicologia y Educacion de la Universidad Veracruzana 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.)

This study proposes a system that applies artificial intelligence tools to reduce the number of vehicular accidents produced by driver drowsiness and distraction. The proposed system, which is suitable for any kind of vehicle, detects the driver’s face, which is fed into a deep neural network that analyzes it. The network output is then fed into a detection stage which determines whether the driver is drowsy or distracted, activating an alarm. The proposed system has a low computational complexity, allowing real time implementation on mobile devices. The experimental results show that the proposed system can detect drowsiness and distraction with an accuracy of 95.8% in high performance computers. It also shows an 84.5% accuracy on mobile devices with limited capacity, keeping an acceptable operational speed for its implementation in real time. It is concluded that the proposed system detects driver drowsiness and distraction with high accuracy. [ABSTRACT FROM AUTHOR]

Copyright of Cirugía y Cirujanos is the property of Publicidad Permanyer SLU 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.)

Copyright of Ingenius, Revista Ciencia y Tecnología is the property of Universidad Politecnica Salesiana 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.)

Copyright of Ingenius, Revista Ciencia y Tecnología is the property of Universidad Politecnica Salesiana 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.)

Copyright of Cuadernos.info is the property of Pontificia Universidad Catolica de Chile, Facultad de Comunicaciones 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.)

Copyright of Revista Pediatría de Atención Primaria is the property of LUA Ediciones 3.0 S.L. 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.)

In the present work, mono and multicomponent adsorption systems of heavy metals (Hg2+, Cu2+ y Ni2+) as adsorbates and carbon nanotubes (CNT) as adsorbents were studied. First, the thermodynamic and QSAR properties at 298.15 and 30815K were determined using computational simulation. Subsequently, Feedforward backprop and Elman backprop artificial neural networks were developed, where the network with the highest precision of the thermodynamic and QSAR properties was the Elman Backprop with the Logsig function using 5 and 3 neurons in the hidden layer at 298.15 and 308.15 K, finally, the networks had an r2 of 0.999, and a mean square error of 0.021, 0.024 and 0.214 respectively. [ABSTRACT FROM AUTHOR]

Copyright of DYNA - Ingeniería e Industria is the property of Publicaciones Dyna SL 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.)

Copyright of ESPOCH Congresses: The Ecuadorian Journal of S.T.E.A.M. is the property of Knowledge E DMCC 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.)

The article "Joint Health Command (JECOSAN)" presents a study on the development of intelligent systems to predict In-hospital Mortality (MIH) and Length of Stay (LGE) in patients with physical trauma. The authors propose and evaluate a set of deep neural networks (DNN) to predict MIH, using clinical characteristics, triage scale, procedure codes, and the International Classification of Diseases (ICD). The results show that DNN outperforms traditional machine learning methods in terms of accuracy, specificity, and sensitivity. From the perspective of operational military health, these methods could be useful for classifying and predicting mortality in traumatized patients in a military environment. Additionally, the importance of including ethics in the curricula of military healthcare personnel is highlighted. [Extracted from the article]

Copyright of Revista Electrónica Interuniversitaria De Formación del Profesorado is the property of Asociacion Universitaria de Formacion del Profesorado (AUFOP) 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.)

Copyright of Revista de la Academia Colombiana de Ciencias Exactas, Físicas y Naturales is the property of Academia Colombiana de Ciencias Exactas, Fisicas y Naturales 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.)

Copyright of ESPOCH Congresses: The Ecuadorian Journal of S.T.E.A.M. is the property of Knowledge E DMCC 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.)

In order to detect potato crop pests from symptom images, a web platform was developed to facilitate the diagnosis of early and late blight from images of affected leaves. The Python programming language, the Flask framework, Bootstrap for platform styling, the necessary libraries, and Pycharm as development environment were used. Image recognition and artificial intelligence software tools, specifically artificial neural networks, were used to implement the diagnostic module. The system also includes the field monitoring record with the incidences of pests, allows incorporating new data, and provides information on management in correspondence with climate variability. For the diagnosis validation, images of symptoms confirmed by experts were taken. The neural network model used was 94.6 % accurate. Future work will include the detection of viruses, bacteria and other fungi from images. The tool is a useful and novel application at the service of agri-food safety, serving as a coordinated platform for plant protection technicians and farmers involved in potato cultivation. [ABSTRACT FROM AUTHOR]

The greatest problems that has been presented in the world, especially in Colombia, is due to an accelerated population growth generated by the overwhelming increase in migrant circulation or mobility of people due to other causes or eventual situations of nature. These social situations have been significantly affecting the urban order of cities, particularly due to the confinement of the population and the increase in environmental pollution that this increase brings, in addition to unhealthiness, poverty, among other factors. To study population growth based on demographic variables and environmental factors, an Artificial Neural Network was built to carry out a data analysis that incorporated related variables such as CO2 and other pollutants. The purpose was to identify the influence of these toxic agents with the growth of a population. Once the dynamic prediction was verified through the neural network, a mathematical model was built to study a specific case of demographic behavior for a certain Colombian region by means of a Forrester Diagram, where the migratory conditions, birth rate, morbidity, mortality, migratory flow (or mobility of people due to other natural events) and pollutants that affect human health mainly in respiratory diseases. The results delivered predictions and the Forrester Diagram confirming the relationship between toxic agents and demographic aspects. [ABSTRACT FROM AUTHOR]

Because of the impact on productivity and cost reduction, decision making in industrial processes is one of the most required aspects in the industry. Specifically in the panel industries, product quality depends on multiple variables, especially wood variability. Among other factors, quality depends on the adhesion of veneers or perpendicular tensile strength. The main objective of this study was to evaluate a Machine Learning approach to predict the adhesion under industrial conditions in the gluing and pre-pressing stage. The control variables that determine this adhesion are mainly: operational times, amount of adhesive, environmental conditions, and veneer temperature. Using Knowledge Discovery in Databases data analytics methodology, Artificial Neural Networks and Support Vector Machine were evaluated. Sigmoid activation function was used with 3 hidden layers and 245 neurons. In addition to the Adam optimizer, Multi-Layer Perceptron, Artificial Neural Networks delivered the best accuracy levels of over 66 %. Best result with Relu and Sigmoid functions were obtained. Sigmoid showed accuracy over 66 %, precision fit good to find positive results (70 %). Relu function obtained the best recall (over 74 %) showing a good capacity to identify reality. Results show that it is not sufficient to generate a data set using the averages of each process variable, since it is difficult to obtain better results with the algorithms evaluated. This work contributes to defining a methodology to be used in plywood plants using industrial data to train and validate Machine Learning models. [ABSTRACT FROM AUTHOR]

Many studies have shown the need in the Venezuelan oil industry to implement viable alternatives in the field of pipeline integrity management. Thus, the aim of this work is to propose a prediction model for the external corrosion rate of burned transmission pipelines in a crude oil production field with clay soils, Incated in the west of Zuha State in Venezuela. After the collection, revision and classification of soil and operating parameter data in the field, a definition of input and output variables was carried out, used to generate 2 models, one regression type and the other classification type. For the neural network model, a low regression fit (R) of 6.62% and an RMSE of 2.13 were obtained, indicators of low model efficiency due to the restrictions of the data provided and sample stre. On the contrary, for the decision tree classification model, un accuracy of 98 141% was obtained, when classifying the corusion rate in seventy ranges. This clasification tree model will serve as a starting point for subsequent research to dleve deeper into the area. [ABSTRACT FROM AUTHOR]

This article presents the development of a prototype computational system designed for the detection of vital signs and the location of people in structures, such as in earthquake situations. The system uses sensor technologies, signal processing algorithms and neural networks to identify the presence of trapped people and monitor their vital signs. The system architecture, component integration, and initial test results in simulated environments are detailed. Preliminary results demonstrate the effectiveness of the system in accurate and real-time detection, proposing a valuable tool for rescue operations. [ABSTRACT FROM AUTHOR]

The objective of this work was the application of an artificial neural network in prediction of holdup of two-phase flow (oil-water) in a pipe in horizontal position. To this end, the velocity superficial of water and oil were used as input parameters, meanwhile, the holdups of these two fluids were used as output parameters for the training and testing of the multilayer neural network, the method used was back-propagation. The experimental data (92 data) were taken at LEMI-EESC-USP and were used to develop the artificial neural network model. Finally, it was concluded that the experimental data used in the neural network agreed with the tagsig transfer function with 10 neurons in the hidden layer evaluated from the absolute percentage error of (AAPE= 3,95) and coefficient of determination (R²= 0,975). [ABSTRACT FROM AUTHOR]

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.)

Copyright of Revista Técnica Energía is the property of Centro Nacional de Control de Energia CENACE 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.)

Copyright of Revista CienciaUAT is the property of Universidad Autonoma de Tamaulipas 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.)

Copyright of Revista Salud UIS 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.)

Copyright of Rect@: Revista Electrónica de Comunicaciones y Trabajos de ASEPUMA is the property of ASEPUMA Asociacion Espanola de Profesores Universitarios de Matematicas 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.)

Copyright of Rect@: Revista Electrónica de Comunicaciones y Trabajos de ASEPUMA is the property of ASEPUMA Asociacion Espanola de Profesores Universitarios de Matematicas 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.)

Copyright of Propiedad Inmaterial is the property of Universidad Externado de Colombia, Centro de Estudios de la Propriedad Intelectual 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.)

Copyright of TECHNO Review is the property of Eurasia Academic Publishing Group 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.)

The main objective of the present research study was to predict the permeability of concrete with mineral ultrafine additions by using artificial neural networks. Concrete mixtures were prepared without addition (4) and with additions of metakaolin (1), microsilica (2), and pozzolan (3). The results showed that permeability values ranged between [0.12, 733]x10-13 m/s. Using the ordinary analysis of variances, it was established that at 28 days the permeability of concrete stabilizes with the additions of (2) and (3), but not with the addition of (1). Different artificial neural network architectures were tested, obtaining correlations in the simulation phase from 0.904 to 0.907. In addition, neural networks were obtained that made feasible to safely predict permeability coefficients for added concrete. It is concluded that additions (2) and (3) have an inverse effect on the permeability of concrete, reaching stability after 28 days. [ABSTRACT FROM AUTHOR]

Copyright of Revista Cubana de Ciencias Informáticas is the property of Universidad de las Ciencias Informaticas (UCI) 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.)

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.)

Copyright of Tecnura is the property of Tecnura 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.)

The Nickel and Cobalt Production Company "Comandante Ernesto Che Guevara" plays an essential role in the economic development of Cuba. Within the industrial production process, the lateritic ore reduction furnace is a key stage and requires a high consumption of fuel oil, which represents around 40% of the unit cost per ton produced of nickel plus cobalt, making it vital. reduce such costs. The implementation of the automatic control system in postcombustion has been affected by not having mathematical models of this process. In the present investigation, a black box model based on artificial neural networks that reflects the dynamic characteristics of the process is proposed. The methodology used consists of carrying out passive experiments of 10-months operating time period for the preparation of a recurrent algorithm that includes the methods of random cross validation and the Akaike's information criterion. As a result, a NARX model (non-linear autoregressive structure with external inputs) was obtained that predicts the temperatures of hearths four and six, obtaining a mean square error of less than 5 °C, with a prediction horizon one step ahead (120s). The accuracy achieved with the model contributes to predicting the thermal profile in the heating zone of the furnace, as a basis for the design of control strategies that guarantee a better use of energy and of the reducing additive fuel, which would reduce process losses. and environmental pollution. [ABSTRACT FROM AUTHOR]