Repositorio Universidad Internacional Iberoamericana
Not a member yet
907 research outputs found
Sort by
Contextual Urdu Lemmatization Using Recurrent Neural Network Models
No full textIn the field of natural language processing, machine translation is a colossally developing research area that helps humans communicate more effectively by bridging the linguistic gap. In machine translation, normalization and morphological analyses are the first and perhaps the most important modules for information retrieval (IR). To build a morphological analyzer, or to complete the normalization process, it is important to extract the correct root out of different words. Stemming and lemmatization are techniques commonly used to find the correct root words in a language. However, a few studies on IR systems for the Urdu language have shown that lemmatization is more effective than stemming due to infixes found in Urdu words. This paper presents a lemmatization algorithm based on recurrent neural network models for the Urdu language. However, lemmatization techniques for resource-scarce languages such as Urdu are not very common. The proposed model is trained and tested on two datasets, namely, the Urdu Monolingual Corpus (UMC) and the Universal Dependencies Corpus of Urdu (UDU). The datasets are lemmatized with the help of recurrent neural network models. The Word2Vec model and edit trees are used to generate semantic and syntactic embedding. Bidirectional long short-term memory (BiLSTM), bidirectional gated recurrent unit (BiGRU), bidirectional gated recurrent neural network (BiGRNN), and attention-free encoder–decoder (AFED) models are trained under defined hyperparameters. Experimental results show that the attention-free encoder-decoder model achieves an accuracy, precision, recall, and F-score of 0.96, 0.95, 0.95, and 0.95, respectively, and outperforms existing model
Deep Learning-Based Multiclass Instance Segmentation for Dental Lesion Detection
No full textAutomated dental imaging interpretation is one of the most prolific areas of research using artificial intelligence. X-ray imaging systems have enabled dental clinicians to identify dental diseases. However, the manual process of dental disease assessment is tedious and error-prone when diagnosed by inexperienced dentists. Thus, researchers have employed different advanced computer vision techniques, as well as machine and deep learning models for dental disease diagnoses using X-ray imagery. In this regard, a lightweight Mask-RCNN model is proposed for periapical disease detection. The proposed model is constructed in two parts: a lightweight modified MobileNet-v2 backbone and region-based network (RPN) are proposed for periapical disease localization on a small dataset. To measure the effectiveness of the proposed model, the lightweight Mask-RCNN is evaluated on a custom annotated dataset comprising images of five different types of periapical lesions. The results reveal that the model can detect and localize periapical lesions with an overall accuracy of 94%, a mean average precision of 85%, and a mean insection over a union of 71.0%. The proposed model improves the detection, classification, and localization accuracy significantly using a smaller number of images compared to existing methods and outperforms state-of-the-art approache
How much do Latin American medical students know about radiology? Latin-American multicenter cross-sectional study
No full textBackground
Radiology is a useful tool for diagnosis and intervention in medical practice, and all the components within the teaching-learning process of this subject during undergraduate studies influence successful knowledge application.
Objective
This study aimed to describe the level of knowledge in radiology of students in the last two years of medical school and curricular characteristics of their courses in seven Latin American countries.
Methods
A multicenter cross-sectional study was carried out on medical students of 7 Latin American countries (Bolivia, Brazil, Colombia, Ecuador, Mexico, Paraguay, and Peru) in their final two years of medical school, using an online questionnaire validated by experts and adapted for each country that assessed knowledge and curricular characteristics in radiology subject. Scores were assigned according to the number of correct answers for the knowledge test. The T-test, and regression analysis with one-way ANOVA were used to search for relationships between the level of knowledge and other variables.
Results
A total of 1514 medical students participated in this study. All countries had similar participation (n > 200); most participants were women 57.8%. The country with the highest knowledge score was Brazil. Male, sixth year (internship) and from public universities students had higher knowledge score (n < 0.05). Participants, who considered radiology more important, and who reported higher compliance with teaching staff with the proposed syllabus, and programmed classes, obtained better scores (n < 0.05).
Conclusions
Latin American medical students included in this study have a regular overall level of knowledge of Radiology, apparently influenced by curricular differences such as class and academic program compliance. Efforts to better understand and improve academic training are indispensable.
Limitations
The study was subject to selection bias determined by non-probability convenience sampling. The questionnaire assessed only theoretical knowledge and the evaluation system was designed by the investigators
Análisis crítico sobre el perfil de salida del bachillerato ecuatoriano. Una mirada desde el método de aprendizaje basado en proyectos
No full textLos programas educativos cada vez más se inclinan a la potenciación de valores que favorezcan el desarrollo integral de los educandos, para ello se implementan diversas fórmulas que pretenden desde lo metodológico ajustarse a las exigencias sociales, educativas y curriculares. En este acercamiento a la formación del Bachiller Ecuatoriano, se analizan sus principios legales, lineamientos curriculares y estándares de calidad educativa enfocado al cumplimiento del perfil de salida del bachillerato, así como la percepción de estos por parte de estudiantes y docentes de la Unidad Educativa del Milenio Manuel J. Calle de la ciudad de Cuenca, a partir de aquí se propone una estrategia de mejora con el uso del Método de Aprendizaje Basado en Proyectos (ABP), aplicada en una muestra de 92 estudiantes del 2do año del Bachillerato General Unificado (BGU), quienes cursaron el Programa de Participación Estudiantil (PPE), específicamente el PPE (2017-2018), cuyos resultados evidencian que el Método ABP empleado en el PPE caso de estudio contribuye significativamente a elevar la calidad del Perfil de Salida del Bachiller (PSB) por medio del desarrollo de habilidades para la vida. El Método de Aprendizaje Basado en Proyectos ABP es una alternativa adecuada para elevar el proceso formativo del país, a la vez facilita la convivencia armónica en el marco escolar para quienes la utilizan directa e indirectamente
Hybridizing Artificial Intelligence Algorithms for Forecasting of Sediment Load with Multi-Objective Optimization
No full textForecasting of sediment load (SL) is essential for reservoir operations, design of water resource structures, risk management, water resource planning and for preventing natural disasters in the river basin systems. Direct measurement of SL is difficult, labour intensive, and expensive. The development of an accurate and reliable model for forecasting the SL is required. Sediment transport is highly non-linear and is influenced by a variety of factors. Forecasting of the SL using various conventional methods is not highly accurate because of the association of various complex phenomena. In this study, major key factors such as rock type (RT), relief (R), rainfall (RF), water discharge (WD), temperature (T), catchment area (CA), and SL are recognized in developing the one-step-ahead SL forecasting model in the Mahanadi River (MR), which is among India’s largest rivers. Artificial neural networks (ANN) in conjunction with multi-objective genetic algorithm (ANN-MOGA)-based forecasting models were developed for forecasting the SL in the MR. The ANN-MOGA model was employed to optimize the two competing objective functions (bias and error variance) with simultaneous optimization of all associated ANN parameters. The performances of the proposed novel model were finally compared to other existing methods to verify the forecasting capability of the model. The ANN-MOGA model improved the performance by 12.81% and 10.19% compared to traditional AR and MAR regression models, respectively. The results suggested that hybrid ANN-MOGA models outperform traditional autoregressive and multivariate autoregressive forecasting models. Overall, hybrid ANN-MOGA intelligent techniques are recommended for the forecasting of SL in river
IoTTPS: Ensemble RKSVM Model-Based Internet of Things Threat Protection System
No full textAn Internet of Things (IoT) network is prone to many ways of threatening individuals. IoT sensors are lightweight, lack complicated security protocols, and face threats to privacy and confidentiality. Hackers can attack the IoT network and access personal information and confidential data for blackmailing, and negatively manipulate data. This study aims to propose an IoT threat protection system (IoTTPS) to protect the IoT network from threats using an ensemble model RKSVM, comprising a random forest (RF), K nearest neighbor (KNN), and support vector machine (SVM) model. The software-defined networks (SDN)-based IoT network datasets such as KDD cup 99, NSL-KDD, and CICIDS are used for threat detection based on machine learning. The experimental phase is conducted by using a decision tree (DT), logistic regression (LR), Naive Bayes (NB), RF, SVM, gradient boosting machine (GBM), KNN, and the proposed ensemble RKSVM model. Furthermore, performance is optimized by adding a grid search hyperparameter optimization technique with K-Fold cross-validation. As well as the NSL-KDD dataset, two other datasets, KDD and CIC-IDS 2017, are used to validate the performance. Classification accuracies of 99.7%, 99.3%, 99.7%, and 97.8% are obtained for DoS, Probe, U2R, and R2L attacks using the proposed ensemble RKSVM model using grid search and cross-fold validation. Experimental results demonstrate the superior performance of the proposed model for IoT threat detection
Molecular epidemiology, transmission and clinical features of 2022‐mpox outbreak: A systematic review
No full textBackground and Aims
The 2022-mpox outbreak has spread worldwide in a short time. Integrated knowledge of the epidemiology, clinical characteristics, and transmission of mpox are limited. This systematic review of peer-reviewed articles and gray literature was conducted to shed light on the epidemiology, clinical features, and transmission of 2022-mpox outbreak.
Methods
We identified 45 peer-reviewed manuscripts for data analysis. The standards of the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) Statement and Cochrane Collaboration were followed for conducting the study.
Results
The case number of mpox has increased about 100 times worldwide. About 99% of the cases in 2022 outbreak was from non-endemic regions. Men (70%–98% cases) were mostly infected with homosexual and bisexual behavior (30%–60%). The ages of the infected people ranged between 30 and 40 years. The presence of HIV and sexually transmitted infections among 30%–60% of cases were reported. Human-to-human transmission via direct contact and different body fluids were involved in the majority of the cases (90%–100%). Lesions in genitals, perianal, and anogenital areas were more prevalent. Unusually, pharyngitis (15%–40%) and proctitis (20%–40%) were more common during 2022 outbreak than pre-2022 outbreaks. Brincidofovir is approved for the treatment of smallpox by FDA (USA). Two vaccines, including JYNNEOSTM and ACAM2000®, are approved and used for pre- and post-prophylaxis in cases. About 100% of the cases in non-endemic regions were associated with isolates of IIb clade with a divergence of 0.0018–0.0035. Isolates from B.1 lineage were the most predominant followed by B.1.2 and B.1.10.
Conclusion
This study will add integrated knowledge of the epidemiology, clinical features, and transmission of mpox
Distributed Denial of Service Attack Detection in Network Traffic Using Deep Learning Algorithm
No full textInternet security is a major concern these days due to the increasing demand for information technology (IT)-based platforms and cloud computing. With its expansion, the Internet has been facing various types of attacks. Viruses, denial of service (DoS) attacks, distributed DoS (DDoS) attacks, code injection attacks, and spoofing are the most common types of attacks in the modern era. Due to the expansion of IT, the volume and severity of network attacks have been increasing lately. DoS and DDoS are the most frequently reported network traffic attacks. Traditional solutions such as intrusion detection systems and firewalls cannot detect complex DDoS and DoS attacks. With the integration of artificial intelligence-based machine learning and deep learning methods, several novel approaches have been presented for DoS and DDoS detection. In particular, deep learning models have played a crucial role in detecting DDoS attacks due to their exceptional performance. This study adopts deep learning models including recurrent neural network (RNN), long short-term memory (LSTM), and gradient recurrent unit (GRU) to detect DDoS attacks on the most recent dataset, CICDDoS2019, and a comparative analysis is conducted with the CICIDS2017 dataset. The comparative analysis contributes to the development of a competent and accurate method for detecting DDoS attacks with reduced execution time and complexity. The experimental results demonstrate that models perform equally well on the CICDDoS2019 dataset with an accuracy score of 0.99, but there is a difference in execution time, with GRU showing less execution time than those of RNN and LSTM
Inclusión Educativa y la Enseñanza Bilingüe en el Nivel de Primaria en un Centro Escolar de Honduras
No full textLa investigación analiza la implementación de un modelo multinivel inclusivo utilizado para la enseñanza del inglés como segunda lengua. Se estudia el modelo MTSS, Sistema de Soporte Multi Nivel, en la clase de lectura en inglés y el rendimiento académico de estudiantes sin y con algún tipo de Necesidad Educativa Especial (NEE) dentro del sistema de educación bilingüe. Se diseñó y se validó un Plan Integral Educativo (PIE) basado en el Index de Inclusión de Ainscow y Booth (2015) y anclado en la estructura multinivel. Específicamente busca examinar la implementación de las prácticas y valores inclusivos en el área de primaria vinculada con la enseñanza bilingüe del inglés como segunda lengua, el aprendizaje y el rendimiento académico. Es un diseño cuantitativo, de modalidad descriptiva y comparativa entre grupo experimental y grupo control. En este, participaron un total de 132 estudiantes del primer grado de primaria de la Escuela Internacional Sampedrana de Honduras, un total de 20 estudiantes con NEE. La recolección de datos se realizó mediante cuestionarios, pruebas académicas específicas de lectura en inglés, observaciones vivenciales y encuesta de satisfacción del grupo experimental. Los datos estadísticos fueron informatizados en los softwares de Excel. Entre los hallazgos, la implementación de PIE y la estructura inclusiva multinivel favorece el rendimiento académico de la lectura en inglés de los estudiantes sin y con NEE. En todas las pruebas académicas los resultados del grupo experimental son superiores al grupo control. El alumnado con algún tipo de NEE logró aprender a leer y comprender en inglés como su segunda lengua. El equipo docente propició el ambiente educativo respetando los ritmos de aprendizaje, tiempo de internalización e implementando acomodaciones curriculares. Las estrategias de apoyo y las prácticas inclusivas dan oportunidades efectivas de aprendizaje de una segunda lengua
Blockchain-Modeled Edge-Computing-Based Smart Home Monitoring System with Energy Usage Prediction
No full textInternet of Things (IoT) has made significant strides in energy management systems recently. Due to the continually increasing cost of energy, supply–demand disparities, and rising carbon footprints, the need for smart homes for monitoring, managing, and conserving energy has increased. In IoT-based systems, device data are delivered to the network edge before being stored in the fog or cloud for further transactions. This raises worries about the data’s security, privacy, and veracity. It is vital to monitor who accesses and updates this information to protect IoT end-users linked to IoT devices. Smart meters are installed in smart homes and are susceptible to numerous cyber attacks. Access to IoT devices and related data must be secured to prevent misuse and protect IoT users’ privacy. The purpose of this research was to design a blockchain-based edge computing method for securing the smart home system, in conjunction with machine learning techniques, in order to construct a secure smart home system with energy usage prediction and user profiling. The research proposes a blockchain-based smart home system that can continuously monitor IoT-enabled smart home appliances such as smart microwaves, dishwashers, furnaces, and refrigerators, among others. An approach based on machine learning was utilized to train the auto-regressive integrated moving average (ARIMA) model for energy usage prediction, which is provided in the user’s wallet, to estimate energy consumption and maintain user profiles. The model was tested using the moving average statistical model, the ARIMA model, and the deep-learning-based long short-term memory (LSTM) model on a dataset of smart-home-based energy usage under changing weather conditions. The findings of the analysis reveal that the LSTM model accurately forecasts the energy usage of smart homes