Ingeniería en Sistemas, Electrónica e Industrial
Permanent URI for this communityhttp://repositorio.uta.edu.ec/handle/123456789/1
Browse
14 results
Search Results
Item Sistema de detección de intrusos (IDS) basado en machine learning para el control de la red en la Unidad Educativa “19 de Septiembre” en la Ciudad de Salcedo(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Tecnologías de la Información, 2025-02) Lascano Banshuy Jairo Guillermo; Sánchez Zumba Andrea PatriciaThe increasing sophistication of cyber threats highlights the need for robust solutions to protect institutional networks. This project focuses on the implementation of a Machine Learning-based Intrusion Detection System (IDS) for the "19 de Septiembre" Educational Unit. Initial network assessments revealed critical security vulnerabilities, including frequent connectivity issues, limited access, and user dissatisfaction, primarily caused by the lack of attention to cybersecurity within the institution. To address these challenges, a hybrid dataset was developed by combining real-time data collected through the IDS with publicly available datasets. This approach ensured the dataset's relevance and robustness, enhancing the model's accuracy in classifying benign traffic and detecting potential attacks. After evaluating various machine learning algorithms, Random Forest was selected due to its high adaptability, strong performance, and compatibility with the project's resource constraints. The trained model achieved exceptional results in terms of accuracy, recall, and overall reliability, providing a solid foundation for the IDS. This approach allows educational institutions and other organizational environments to proactively adapt to emerging cyber threats, which are constantly evolving in complexity and scope. By strengthening the network infrastructure, a secure and reliable environment is fostered, protecting both data and the institution's critical operations. Furthermore, this improvement contributes to ensuring user trust in the overall use of the infrastructureItem Sistema informático aplicando python para la gestión de productos con series de tiempo en la panadería y pastelería “Flor de Cebada"(2025-02) Freire Valencia Jean Carlo; Aldas Flores Clay Fernando; Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Tecnologías de la InformaciónThe present project focuses on the development of a software system aimed at improving product management processes in the bakery and pastry shop "Flor de Cebada." Leveraging Python, the Flask framework for the backend, and React for the frontend, alongside the application of machine learning, the system integrates time series forecasting to enhance trend detection and demand prediction. The goal is to optimize sales operations, minimize product waste, and support data-driven decision-making in a small business environment. The system includes an inventory management mechanism, enabling the registration, updating, and tracking of products, including essential details such as quantities, expiration dates, and prices. Additionally, sales processes are streamlined through barcode scanning, automatic calculations, and totals. A key feature is the generation of reports required by the business. The predictive analysis implemented in the system development provides accurate demand forecasts, helping align production with market needs. The system also generates electronic receipts for transactions, improving operational efficiency and user experience. The project was developed using the Extreme Programming methodology, emphasizing iterative design, stakeholder collaboration, and adaptability to changing requirements. The system delivers tangible benefits for business operations, enabling the bakery to achieve greater accuracy in inventory planning, enhance customer satisfaction, and establish a foundation for sustainable growth. This initiative highlights the potential of technology-driven solutions to address real-world challenges faced by small and medium-sized enterprisesItem Sistema automático de detección de extorsiones mediante el análisis de señales de la voz(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones, 2024-08) Tinta Capuz, Carlos Jair; Castro Martin, Ana PamelaThis project presents the implementation of an automatic system to detect extortion through the analysis of voice signals. This system has been designed to identify extortive language, being an essential tool for security and crime prevention. The architecture of the system is divided into three main layers. In the first, speech recognition and feature extraction is done internally with specific libraries. Extortionist and non-extortionist phrases are classified to train the detection model and ensure accuracy in distinguishing both types of language. A set of sentences classified as extortionate and non-extortionate is incorporated for training. This dataset is used to train the detection model, ensuring that the system can distinguish between the two types of language. The intermediate layer uses advanced natural language processing (NLP) and machine learning techniques with the RandomForestClassifier model, selected for its reliability and efficiency. This system processes and classifies sentences according to their content, ensuring accurate and robust detection of extortionate language. The presentation layer includes a graphical user interface (GUI) developed with Tkinter. This interface allows users to interact with the system intuitively by activating the microphone to record their voice and receive an immediate assessment of the presence of extortionate language. The GUI improves the accessibility and usability of the system. This prototype represents a significant advance in extortion detection with 92% accuracy according to system tests, providing an efficient solution. It is anticipated that this tool will meet the growing demand for intelligent security systems, helping organizations stay ahead in the fight against crime.Item Sistema inteligente usando video cámaras y motion capture para monitoreo de la rehabilitación de la mano derecha(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Tecnologías de la Información, 2023-09) Bonilla Quishpe, Miguel Enrique; Nogales Portero, Rubén EduardoRehabilitation monitoring has become a prominent advancement in the technological field, thanks to the convergence of different technologies. The integration of these technologies allows for more accurate and detailed monitoring of hand rehabilitation gest ures. Through data capture and analysis real time information about the rehabilitation progress is provided, enabling healthcare professionals and users to assess their performance during the process. Moreover, the system not only offers visual tracking of daily progress but also motivates users to continue their treatment, contributing to the success of their rehabilitation. In this research project, various technologies are implemented to facilitate the monitoring and analysis of hand rehabilitation data . Firstly, the Leap Motion Controller is used to capture hand position and direction data during the performed gestures. These data are processed using a custom acquisition system developed in Matlab, enabling their analysis, and generating a dataset. The dataset is preprocessed and relevant features are extracted for the classification of each gesture. For this, different models are implemented, such as Artificial Neural Networks (ANN) and K Nearest Neighbors (KNN) to obtain a high accuracy in the classifi cation. After training, ANN with 93% accuracy was selected. Once the model is trained, it is integrated into Unity, where an interactive interface is created to provide visual feedback to the user, allowing them to perform the gestures effectively and moni tor their progress during rehabilitation.Item Aplicación de algoritmos de Machine Learning para predecir la deserción estudiantil en alumnos de primer y segundo semestre en universidades públicas del Ecuador.(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2023) Rodríguez Vásconez, Cristóbal Alejandro; Benalcázar Palacios, Marco EnriqueSe estima que en Ecuador la tasa de deserción en los dos primeros semestres de universidad es del 20%. Existen factores socioeconómicos que influyen en el abandono académico de un estudiante. La carencia de programas que atiendan la insatisfacción estudiantil provoca que no se detecten problemas a tiempo y no se puedan aplicar acciones correctivas oportunamente. En este proyecto se aplican técnicas de Machine Learning para predecir la deserción estudiantil a partir de factores seleccionados: socioeconómicos, psicológicos, demográficos y académicos. Partimos de la recolección y tratamiento de datos y se usaron Redes Neuronales Artificiales para crear un modelo que clasifica a un estudiante entre desertor o a salvo de deserción. Se evalúan las métricas Acurracy, sensibilidad y especificidad para determinar qué tan eficiente es el modelo. El modelo final es capaz de clasificar estudiantes a salvo de deserción de forma correcta el 87% de las veces y logra clasificar a desertores de forma correcta el 60% de las veces.Item Modelo de Machine Learning para mitigar los fraudes informáticos de phishing basados en la ingeniería social en la Facultad de Ingeniería en Sistemas Electrónica e Industrial(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Tecnologías de la Información, 2023-03) Jaramillo Basantes, Fabiana Patricia; Nogales Portero, Rubén EduardoNowadays, Phishing websites continue to be a significant threat in the vast cyberspace of the internet. When a user visits a Phishing URL, attackers obtain the user's personal and confidential information. Cyber criminals use various social engineering techniques to carry out identity theft or launch targeted attacks. Students and professors are not exempt from the strong influence of different social engineering techniques. Phishers seek ways to harm and make money through the manipulation and extortion of unsuspecting users. To address this problem, the present curricular integration work proposes implementing a Machine Learning model for Phishing detection deployed as a browser extension. 24 characteristics of its structure were extracted for the analysis of URLs and construction of the dataset. A comparison was made between various classification models such as Random Forest (RF), Support Vector Machine (SVM), Artificial Neural Network (RNA), and K-Nearest Neighbors (KNN) to choose the most appropriate and best fit for the problem. Once the algorithms of the different training models were analyzed, the model used to classify URLs is an artificial neural network (RNA) achieving an accuracy of 99.98%. The purpose of this work is to help the FISEI community. The extension will mitigate and prevent users from becoming victims of malicious activities such as falling into Phishing URLs that apply various principles of Social Engineering.Item Sistema electrónico de notificación de emergencias basado en IoT para la asistencia médica a personas de la tercera edad.(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Ingeniería en Electrónica y Comunicaciones, 2023-03) Guamán Egas, Jefferson Ismael; Castro Martin, Ana PamelaIn the present work, a system with IoT architecture is implemented that allows detecting and alerting falls in older adults, these events being of great incidence in this group of people; In addition, vital signs are monitored. A bracelet type device was designed with the ESP32 development board with the MPU6050 sensor for fall detection and the MAX30100 and MLX90614 sensors for monitoring heart rate, oxygen saturation and body temperature. For fall detection, models based on "Machine Learning" are created using the "open source" platform Edge Impulse. To create the model, 10 participants were drawn: 5 men and 5 women. Each participant made 5 types of falls, each fall was repeated 10 times, obtaining 50 for each participant, in a similar way activities of daily living carried out by an elderly person were implemented. For the mobile application, Flutter compatible with Android and IOS was brought, different widgets were programmed for the development of the graphical interface to build sensor data, user information, create alerts on the status of vital signs at an interval of time and to present the history of alerts. Dependencies such as "provider" were used to manage the state of the application, allowing control of connectivity and incoming messages through the MQTT protocol. To alert the caregiver of the elderly, push notifications based on Google's own "Firebase Cloud Messaging" architecture are used. An IoT device was obtained based on a microcontroller with the ability to: connect to the WiFi network to send data to a server; run the "Machine Learning" model for fall detection; monitor heart rate, oxygen saturation and body temperature through the wrist; and generate automatic alerts on the cell phone through the application.Item Asistente electrónico de agricultura hidropónica aplicando Machine Learning(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Ingeniería en Electrónica y Comunicaciones, 2023-01) Solis Salazar, Juan Sebastian; Ayala Baño, Paulina ElizabethThis research paper details the implementation of an electronic assistant for hydroponic agriculture applying machine learning. The assistant was implemented in the Puerto Arturo Sector an NFT (Nutrient Film Technique) system; The wizard allows you to control the ranges of pH, Total Dissolved Solids (TDS) and Electrical Conductivity of the nutrient solution of the hydroponic system with a command issued by the "K Nearby Neighbors" algorithm configured on a cloud server, in order to provide constant and precise control of nutrient solution rates for leafy vegetable crops. The assistant has two electronic boards with Wi-Fi that connect to the local area network of the greenhouse to send and receive the necessary data for the supervision and control of the nutrient solution of the NFT system. The assistant uses the MQTT (Message Queue Telemetry Transport) protocol that allows communication between the electronic boards, facilitates the creation of a server in the cloud and allows the configuration of the machine learning algorithm to use it as a command for the assistant's actuators. In addition, it has a mobile and web interface that allows monitoring the data and selecting the type of control of the nutrient solution, on the other hand, the assistant sends alerts when the monitoring ranges change to atypical values and restricts the activation of the dosing pumps of the NFT system.Item Sistema electrónico de monitoreo de bioseñales para el diagnóstico médico de COVID-19 en personas mediante inteligencia artificial(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones, 2022-09) Gómez Lagua, Santiago Adolfo; Pallo Noroña, Juan PabloThe present research proposes the implementation of an electronic system for the monitoring of vital signs and diagnosis of COVID-19 in patients by means of artificial intelligence. To this end, bibliographic research in medical databases has been used to collect information about the symptoms that occur in patients with COVID-19, this point is important because the artificial intelligence algorithm is more efficient when there is enough data to learn. The dataset is composed of labeled data which are "Ambulatory", "Medical review" and "Hospitalization", so it is chosen to use supervised learning classification algorithms. A data exploration is performed, and then with the vital signs that are acquired from the MAX30100 and MLX90614 sensors, to train and choose the most efficient algorithm. The algorithms used are Support Vector Machines, Decision Trees and Multinomial Naive Bayes. All this process is performed on a Raspberry Pi Zero taking advantage of its resources and dimension suitable for a bracelet, the data are stored in a local database and sent to the cloud via HTTP requests, in order that the treating physician is kept informed of the health status of their patients, so you can also receive alert emails indicating the level of risk of the same. Using the evaluation metrics for the artificial intelligence algorithms, it was determined that the Support Vector Machine algorithm is the one that best fits in the classification of the 3 categories, having as results for the "Ambulatory" class a "Recall" equal to 0.75 and a precision of 0.90, for the "Medical Review" class 0.92 and 0.83 and for the "Hospitalization" class 1.00 and 0.95, concluding that the algorithm efficiently handles the classification of the 3 classes with respect to the other algorithms.Item Evaluación con red neuronal del proceso de corte láser por CO 2 en materiales compuestos de fibra de cabuya(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Física Aplicada, 2022) Almache Barahona, Verónica Carolina; Pérez Salinas, Cristian FabiánThe application of Machine Learning today has allowed the development of learning models to solve problems in different fields of industry. This research work focused on relating neural networks (ANN) with the manufacture of composite materials (polyester matrix + fiber cabuya) and CO2 laser cutting machining. The objective is to develop a neural network to evaluate the application of machine learning to predict the surface finish characteristic of the material under study. The established cutting parameters were laser power and cutting speed. The surface finish characteristic to be evaluated was the surface roughness of the cut composite material. The sheet of the constructed composite material was subjected to CO2 laser cutting, which generated a set of 84 specimens. Experimental data was generated by measuring surface roughness through laboratory tests. The programming of the neural network was done with the Scikit-learn package. This is one of the most widely used open source libraries for machine learning available in Python. The results achieved by the prediction of the network based on the experimental data are related to the values predicted by the neural network model (ANN) and the performance of the network was evaluated using statistical metrics. The statistical results obtained were 0.946, 0.139 and 0.301 corresponding to the coefficient of determination (R2), the mean square error (MSE), and the mean absolute error (MAE) respectively. Therefore, it could be concluded that the performance of the developed neural network has a high validity and ability to predict surface roughness.