Tópicos Sugeridos dentro de su búsqueda.
Tópicos Sugeridos dentro de su búsqueda.
Buscar alternativas:
detection models » detection tools (Expander búsqueda), prevention models (Expander búsqueda), prediction models (Expander búsqueda)
para detection » para deteccion (Expander búsqueda), pcr detection (Expander búsqueda), early detection (Expander búsqueda)
detection models » detection tools (Expander búsqueda), prevention models (Expander búsqueda), prediction models (Expander búsqueda)
para detection » para deteccion (Expander búsqueda), pcr detection (Expander búsqueda), early detection (Expander búsqueda)
1
artículo
Publicado 2024
Enlace
Enlace
The primary focus of this article was to employ Ultralytics technology, specifically YOLOv8, in object recognition. This involved utilizing supervised learning and other machine learning techniques. The article took into consideration the definitions of object detection and model training to effectively categorize solid waste, thereby facilitating recycling efforts. Following this, each object class was manually identified using the LabelImg tagger, considering the positions of the objects within the images. This approach led to the analysis of 1517 images and produced notably high-quality and significant results.
2
artículo
This study addresses the global issue of marine pollution, with a particular focus on plastic bag contamination, by leveraging real-time object detection techniques powered by deep learning algorithms. A detailed comparison was carried out between the YOLOv8, YOLO-NAS, and RT-DETR models to assess their effectiveness in detecting plastic waste in underwater environments. The methodology encompassed several key stages, including data preprocessing, model implementation, and training through transfer learning. Evaluation was conducted using a simulated video environment, followed by an in-depth comparison of the results. Performance assessment was based on critical metrics such as mean average precision (mAP), recall, and inference time. The YOLOv8 model achieved an mAP50 of 0.921 on the validation dataset, along with a recall of 0.829 and an inference time of 14.1 milliseconds. The YOLO-N...
3
artículo
This study addresses the global issue of marine pollution, with a particular focus on plastic bag contamination, by leveraging real-time object detection techniques powered by deep learning algorithms. A detailed comparison was carried out between the YOLOv8, YOLO-NAS, and RT-DETR models to assess their effectiveness in detecting plastic waste in underwater environments. The methodology encompassed several key stages, including data preprocessing, model implementation, and training through transfer learning. Evaluation was conducted using a simulated video environment, followed by an in-depth comparison of the results. Performance assessment was based on critical metrics such as mean average precision (mAP), recall, and inference time. The YOLOv8 model achieved an mAP50 of 0.921 on the validation dataset, along with a recall of 0.829 and an inference time of 14.1 milliseconds. The YOLO-N...
4
artículo
Publicado 2021
Enlace
Enlace
The results of 4 predictive models, logistic regression, decision trees, KNN and a neural network are compared to predict the academic dropout of students at the National Intercultural University of the Amazon, applied to a dataset extracted from the system's database. of academic management of the university, which contains socioeconomic and academic performance data which were processed and formatted using onehotencoding techniques in order to apply the predictive models already mentioned. For data processing and formatting, Transac Sql queries were used and the application of predictive models was done through Knime Software and using Python through Google Colab. The results obtained by applying 4 predictive models are very good since they all exceeded 80% of Accuracy, which guarantees that they can be put into production for the benefit of the university and thus can make better deci...
5
tesis de grado
Publicado 2020
Enlace
Enlace
Cyberbullying is a social problem in which bullies’ actions are more harmful than in traditional forms of bullying as they have the power to repeatedly humiliate the victim in front of an entire community through social media. Nowadays, multiple works aim at detecting acts of cyberbullying via the analysis of texts in social media publications written in one or more languages; however, few investigations target the cyberbullying detection in the Spanish language. In this work, we aim to compare four traditional supervised machine learning methods performances in detecting cyberbullying via the identification of four cyberbullying-related categories on Twitter posts written in the Peruvian Spanish language. Specifically, we trained and tested the Naive Bayes, Multinomial Logistic Regression, Support Vector Machines, and Random Forest classifiers upon a manually annotated dataset with th...
6
tesis de grado
Publicado 2020
Enlace
Enlace
Botnets are some of the most recurrent cyber-threats, which take advantage of the wide heterogeneity of endpoint devices at the Edge of the emerging communication environments for enabling the malicious enforcement of fraud and other adversarial tactics, including malware, data leaks or denial of service. There have been significant research advances in the development of accurate botnet detection methods underpinned on supervised analysis but assessing the accuracy and performance of such detection methods requires a clear evaluation model in the pursuit of enforcing proper defensive strategies. In order to contribute to the mitigation of botnets, this paper introduces a novel evaluation scheme grounded on supervised machine learning algorithms that enable the detection and discrimination of different botnets families on real operational environments. The proposal relies on observing, u...
7
artículo
Publicado 2025
Enlace
Enlace
La tuberculosis es una enfermedad antigua y que actualmente sigue afectando a nivel mundial. Es la segunda enfermedad infecciosa más mortífera del mundo superada únicamente por la Covid, según la Organización Mundial de la Salud. Debido a esto, se propone un enfoque para la detección tuberculosis a través de imágenes de radiografías de tórax, aplicando dos modelos de Deep learning más utilizadas en la literatura: “Mobilenet” y “Resnet-50” y una “CNN sin arquitectura predefinida”. El enfoque se desarrolla en 4 fases: (1) Adquisición del dataset, (2) desarrollo de los modelos mencionados, (3) la evaluación del desempeño y (4) análisis de resultados. El dataset está conformado por 1,158 imágenes de radiografías de tórax clasificadas en dos clases: “Normal” y “Tuberculosis”. Los resultados evidenciaron que el modelo “Resnet-50” tuvo un mejor rendimi...
8
artículo
Publicado 2021
Enlace
Enlace
Computer vision is one of the fields of Artificial Intelligence that is flourishing because it focuses on the development and improvement of techniques that allow computers to identify, process and classify images, in a way that resembles human vision. This feature makes them an excellent tool for vehicle control systems. For this reason, we developed a system for the recognition of Mexico City license plates using artificial vision techniques, image processing and automatic learning, in order to monitor and speed up response times, when a stolen vehicle is found.
9
artículo
Publicado 2011
Enlace
Enlace
Nowkday the customers requires, complex edifications, it involves complex projects, with a wide variety of materials, installations, resources and procedures that requires the application of administration and planning of high performance, during the construction. Too, is necessary a constant and efficient revision of compatibilities, and feedback in the design of the project, before to begin the construction phase. Frequently the project design phase is completed and the construction phase is begun with a not optimized design, this last involves mistakes, incompatibilities and clashes between the specialties of the engineering. So the responsibility of the final design accord to the normative construction, pass to the builder company, so is common develop solutions and redesigns during the construction phase, this last is a bad and very critical situation, and can affect negatively on h...
10
artículo
Publicado 2011
Enlace
Enlace
Nowkday the customers requires, complex edifications, it involves complex projects, with a wide variety of materials, installations, resources and procedures that requires the application of administration and planning of high performance, during the construction. Too, is necessary a constant and efficient revision of compatibilities, and feedback in the design of the project, before to begin the construction phase. Frequently the project design phase is completed and the construction phase is begun with a not optimized design, this last involves mistakes, incompatibilities and clashes between the specialties of the engineering. So the responsibility of the final design accord to the normative construction, pass to the builder company, so is common develop solutions and redesigns during the construction phase, this last is a bad and very critical situation, and can affect negatively on h...
11
artículo
Publicado 2022
Enlace
Enlace
Computer vision is one of the fields of Artificial Intelligence that is flourishing because it focuses on the development and improvement of techniques that allow computers to identify, process and classify images, in a way that resembles human vision. This feature makes them an excellent tool for vehicle control systems. For this reason, we developed a system for the recognition of Mexico City license plates using artificial vision techniques, image processing and automatic learning, in order to monitor and speed up response times, when a stolen vehicle is found.
12
artículo
Publicado 2020
Enlace
Enlace
Objective: to develop a microfluidic system (lab-on-a-chip) for detecting circulating breast cancer tumor cells. Materials and methods: the device was designed using 3D technology, and it was manufactures using soft photolithography and a laser cutting machine. The system performance and its magnetic settings were assessed using Jurkat cells and breast cancer cells that show different expression of CD45 and EpCAM surface markers. Antibodies against these markers were bound to magnetic pellets. Additionally, iron nanoparticles were used for assessing their entrapment. Results: nanoparticles were significantly trapped in the area set by magnetic field modeling. Tumor cells labeled with magnetic antibodies became trapped. Conclusions: we were able to manufacture a lab-on-a-chip system that is capable to trap circulating breast cancer tumor cells, which may become an excellent tool for diagn...
13
artículo
Publicado 2019
Enlace
Enlace
This study shows the results of investigations to determine harmonics in electrical current through the use of Artificial Neural Networks (ANN) using the methods of Feedforward-Backpropagation through a generator of electrical signals in C# (C Sharp). We studied the causes of current harmonics, what a re its implications in everyday work and filters to attenuate these harmonics. For generation of harmonics, we implemented a transmitter of electrical signals by software, also developed in C# (C Sharp) so as to obtain raw and real data as possible, in order to perform tests for simulating errors in the signal power that occur in real time and then process this data.It was determined that the best method for the detection of harmonics using Artificial Neural Networks is Feedforward — Backpropagation with supervised training in order to handle the input and output to get a better result.Th...
14
artículo
Publicado 1998
Enlace
Enlace
Sintered tin-oxide-based coating have been fabricated for ethanol vapor sensing. After an appropriate annealing process, particles of low coalescence levels were obtained. The vapor detection is based on the surface resistance variation of the coating due to ethanol vapor presenc. A simplified theoretical model based on surface adsorption of ionized molecules was used in order to explain this behavior. Our data fitted quite well in the range of 10-130 ug/ml. For vapor concentration in air we found an experimental value of n=0.63 in the electrical conductance relationship G=G0 [v]m
15
artículo
Publicado 2018
Enlace
Enlace
Globally, high-cost diseases are generating a loss of social efficiency in a tendential manner, different causes are generated by these diseases, but the consequences for social insurance are reflected in unsustainable increases in costs. This work intends to serve as support to determine which diseases can be considered as high cost based on a frequency versus severity analysis, in such a way that it serves as support in the decision making of health policy makers, although the issue of high-cost diseases goes beyond a quantitative-qualitative model, especially for the health of people, this work aims to serve as a basis for a management of health services based on costs that complement the efficient and particular management of all kinds of high cost diseases.
16
objeto de conferencia
The diagnostic process of respiratory diseases requires experience and skills to assess the different pathologies that patients may develop. Unfortunately, the lack of qualified radiologists is a global problem that limits respiratory diseases diagnosis. Therefore, it will be useful to have a tool that minimizes errors and workload, improves efficiency, and speeds up the diagnostic process in order to provide a better healthcare service to the community. This research proposes a methodology to detect pathologies by using deep learning architectures. The present proposal is divided into three types of experiments. The first one evaluates the performance of feature descriptors such as SIFT, SURF, and ORB in medical images with machine learning models as an introduction to the last experiment. The second one evaluates the performance of deep learning architectures such as ResNet50, Alexnet,...
17
tesis doctoral
Publicado 2019
Enlace
Enlace
Esta tesis trata el problema de descubrir valores atípicos temporales interesantes en un conjunto de datos. Presentamos reglas de asociación probabilísticas como medidas para descubrir valores atípicos temporales interesantes basados en el conocimiento del dominio que ha sido aprendido y representado por una Red Bayesiana Dinámica. Las redes Bayesianas dinámicas capturan el conocimiento previo en una relación causal entre variables aleatorias. Las dos reglas de asociación probabilística definidas como: i) soporte bajo & confianza alta y ii) soporte alto & confianza baja, fueron usadas para identificar escenarios donde las discrepancias entre las probabilidades previas y condicionales son significativas. Nuestro enfoque novedoso une ambos métodos y nos permite descubrir valores atípicos temporales interesantes y proporcionan una contextualización en forma de sub-espacios relac...
18
tesis doctoral
Publicado 2024
Enlace
Enlace
Knowing what words of a language are inherited from the ancestor language, which are borrowed from contact languages, which are recently created, and the timing of critical events in the culture, enables modeling of language history including language phylogeny, language contact, and other novel influences on the culture. However, determining which words or forms are borrowed and from whom is a difficult, time consuming, and often fascinating task, usually performed by historical linguists, which is limited by the time and expertise available. While there are semi-automated methods available to identify borrowed words and their word donors, there is still substantial opportunity for improvement. We construct a new language model based monolingual method, competing cross-entropies, based on word source groupings within monolingual wordlists; improve existing multilingual sequence comparis...
19
artículo
Publicado 2023
Enlace
Enlace
The article explores the use of convolutional neural networks, specifically ResNet-50, to detect weevils in corn kernels. Weevils are a major pest of stored maize and can cause significant yield and quality losses. The study found that the ResNet-50 model was able to distinguish with high precision between weevil-infested corn kernels and healthy kernels, achieving values of 0.9464 for precision, 0.9310 for sensitivity, 0.9630 for specificity, 0.9469 for quality index, 0.9470 for the area under the curve (AUC) and 0.9474 for the F-score. The model was able to recognize nine out of ten weevil-free corn kernels using a minimal number of training samples. These results demonstrate the efficiency of the model in the accurate detection of weevil infestation in maize grains. The model's ability to accurately identify weevil-affected grains is critical to taking rapid action to control th...
20
objeto de conferencia
Publicado 2019
Enlace
Enlace
El texto completo de este trabajo no está disponible en el Repositorio Académico UPN por restricciones de la casa editorial donde ha sido publicado.
