Tesis Software

URI permanente para esta colecciónhttp://repositorio.uta.edu.ec/handle/123456789/39538

Examinar

Resultados de la búsqueda

Mostrando 1 - 10 de 135

Implementación de un módulo de pre-validación documental con orquestación dinámica basada en inteligencia artificial generativa para la reducción de expedientes no conformes en el MICC
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software., 2026-07-08)
Tite Haro, Ramiro Matias
;
Fernández Peña, Félix Oscar
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software.
Artículo Académico
Sistema de reconocimiento facial para control de acceso con Deep Learning en la Cooperativa de Ahorro y Crédito Sol en el área de cajas/bóveda.
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software., 2026-07-08)
Yambay Miño, Santiago Estiven
;
Balarezo López, Julio Enrique
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software.
Access control to the cash register and vault areas of Cooperativa de Ahorro y Crédito SOL presented security vulnerabilities due to the reliance on manual logs, which generated risks of identity impersonation, recording errors, and lack of traceability. Therefore, the main objective of this research was to develop an automated access validation application based on facial recognition using convolutional neural networks. The study applied a field and applied methodology with a mixed approach, using surveys, interviews, and observation sheets addressed to a sample of 12 employees in order to diagnose the initial situation. As a result, the proposed application mitigated four high-severity risks and fulfilled seven operational requirements. Furthermore, when comparing two technological models, the transfer learning program implemented with the DeepFace library showed better comparative performance than the local supervised learning classifier, since it used 100 reference photographs per authorized employee, achieving an accuracy of 98 percent, a precision of 96.15 percent, and a sensitivity of 100 percent. In contrast, the local model required 300 photographs and obtained 55 percent accuracy with 89 false negatives. Thus, it was concluded that the implementation of neural networks reduced dependence on physical records and automated biometric verification in real time, providing a more secure, traceable, and auditable access management process for the financial institution.
Sistema de tutoría docente basado en IA generativa supervisada.
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software., 2026-07-08)
Zhu Ordoñez, Erick Daniel
;
Fernández Peña, Félix Oscar
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software.
In higher education, students often receive late or intermittent feedback, and teachers rarely have timely information about their groups’ learning. This work develops a teacher-supervised generative AI tutoring system, which provides students with support aligned to the content of each lesson and gives teachers a consolidated view of their group’s progress. The system exposes a REST API through which the intelligent tutor and the feedback features were integrated into NousAI, which brings together three complementary proposals. The proposal articulates three components: an intelligent tutor contextualized by learning object, weekly feedback for the teacher at the learning-object and module levels, and weekly feedback for the student by module. Supervision rests with the teacher, who decides which learning objects are published and accesses feedback that takes into account students’ actions. Validation applied the System Usability Scale questionnaire to a sample of students and a qualitative questionnaire to five teachers. Perceived usability exceeded the adopted reference value, with a mean of 71.09 out of 100 from 154 valid responses and a 95 % confidence interval between 68.34 and 73.84. Usability proved solid (72.79) against a lower learnability (64.29), which concentrates the greatest room for improvement. Teachers expressed unanimous willingness to keep using it, with confidence in the information generated by the model as the main reservation.
Clasificación de residuos mediante una aplicación android con un modelo entrenado para el aprendizaje en niños de primaria
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software., 2026-07-07)
Peña López, John Alejandro
;
Ibarra Torres, Oscar Fernando
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software.
This research addresses the problem of incorrect waste classification through the development of "MiniEco", an Android mobile application integrated with artificial intelligence for primary school students at the Juan León Mera La Salle Educational Unit. The Mobile-D methodology was used for software development and CRISP-DM for artificial intelligence. After initially training a ResNet50 architecture with TrashNet, it failed in real-time due to unvaried backgrounds and camera frame processing issues. Therefore, the system was optimized using a more diverse, real world dataset and selecting the EfficientNetV2S model. This model was converted to TensorFlow Lite to ensure 100 percent offline operation, achieving a laboratory accuracy of 94.50 percent with a final application size of 36 MB, which maintained a CPU consumption of 16 percent on low-end hardware. In the validation phase with 32 students, a significant increase in ecological knowledge was recorded, moving from 18.75 percent correct answers in the initial diagnosis to an effectiveness of up to 93 percent in tests with the tool. The usability evaluation using the SUS metric showed a satisfaction rate of 96 percent among the students. It is concluded that the integration of local computer vision and mobile gamification strengthens environmental learning at an early age, overcoming institutional connectivity limitations and providing an autonomous, scalable, and highly efficient pedagogical tool.
Sistema automatizado de identificación y clasificación de fotografías de ciclistas en competencias deportivas de descenso utilizando técnicas de visión por computadora.
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software., 2026-07-07)
Villacres Morales, Pablo Martin
;
Fernández Peña, Félix Oscar
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software.
Manual classification of downhill cycling event photographs by bib number and equipment color takes between 2 and 4 days per event in single-operator businesses such as Titan TV, leading to buyer abandonment and constraining business scalability. This work develops an automated computer vision system that reduces that time. Model selection combined the bibliographic review of 30 prior studies, a comparative benchmark of 16 systems over 67 reserved images and 1,290 human judgments, and training on a proprietary dataset of 1,200 photographs from 6 events. The selected models are YOLO11m for rider detection, PARSeq for optical bib recognition and Gemini 2.5 Flash for chromatic analysis. The system is deployed with a NestJS backend, a FastAPI service, a Vue 3 SPA and Cloudflare workers that serve public photographs with watermark and QR code. Operational validation over 4 events processed 2,673 photographs with zero pipeline failures. Automated bib reading covered 85.8 % of the set, with human correction limited to 4.4 % of the total. Calendar time per event dropped from about 96 hours to a range of 60 to 111 minutes, a reduction close to 98 %. Operating cost stood at 23.91 USD per month. The system guarantees complete same-day event delivery and enables scaling of sports photography production without manual work proportional to volume.
Implementación de una aplicación web-móvil con tecnología blockchain para la trazabilidad y gestión de actas de partidos de la Liga Deportiva de Picaihua
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software., 2026-07-07)
Villafuerte Quispe, Jonathan Eduardo
;
Guevara Aulestia, David Omar
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software.
The objective of this research was to implement a web-mobile application based on blockchain technology to provide traceability in the registration, management, and consultation of match reports for the Liga Deportiva Parroquial de Picaihua. During the development of the project, functional, non-functional, and security requirements were analyzed, identifying a total of 61 requirements necessary for the system construction. The proposed solution consisted of an administrative web application and a mobile application for real-time registration of sports reports. In addition, a smart contract was implemented on the Ethereum network, together with cryptographic hash validation mechanisms and automatic auditing processes to strengthen the integrity, authenticity, and traceability of the information. The results obtained showed a significant improvement compared to the traditional process of sports report management. The report submission time was reduced from 24 hours to 40 seconds, while the verification time decreased from approximately 72 hours to a maximum of 1 hour. Likewise, information availability improved from 24 hours to 10 seconds after report validation. The implementation of the system optimized administrative and sports processes, reduced the risk of information falsification, and improved security and reliability in the management of sports reports within the Liga Deportiva Parroquial de Picaihua.
Aplicación móvil para la gestión de adopciones y reporte de mascotas extraviadas con generación de imágenes mediante IA en la Fundación Huellitas
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software., 2026-07-06)
Tonato Gudiño, Alex Ariel
;
Jara Moya, Santiago David
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software.
Animal abandonment and overpopulation place sustained pressure on rescue organizations that still rely on manual procedures. At the Fundacion Huellitas, information about available animals was scattered across social media, paper records reached occupancy near ninety percent, search filters were absent, and adopter evaluation depended on manual judgment without digital validation. This fragmentation limited adoption campaigns and weakened the adopter experience. The study aimed to develop a mobile application that facilitates the adoption of stray animals and improves user experience within the foundation. A field and applied research approach combined surveys with potential adopters and interviews with foundation staff to define functional and non-functional requirements. The application was developed under the Rapid Application Development methodology across its four phases, using React Native with Expo, Cloud Firestore, Firebase Authentication, Storage, Cloud Messaging, and Cloud Functions as an intermediary layer toward the Gemini image-generation model. The system integrated four independent modules: adoption, lost pets, generative artificial intelligence, and administration. Acceptance testing across four iterations on a physical Android device verified twenty-six functional and non-functional criterio with full compliance. Usability was measured through the System Usability Scale applied to twelve potential adopters who first completed five representative tasks, yielding a mean of 80.21 and a standard deviation of 4.05. Every participant exceded the industry benchmark of sixty-eight points, and 66.7 % reached the good-usability threshold. The findings confirm that the application improved the adopter experience over the previous manual process and offer a scalable basis for future animal-welfare features.
Clasificación automática de enfermedades de la papa mediante visión computacional usando android sin Internet
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software., 2026-07-06)
Vichicela Casa, Kevin Daniel
;
Ibarra Torres, Oscar Fernando
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software.
Potato cultivation in the Ecuadorian Andes faces significant losses due to foliar diseases, the rapid identification of which is limited by the scarcity of specialized assistance due to restricted connectivity. This work addresses the question: Is it possible to implement a classification system for multiple phytosanitary diseases affecting potato leaves using computer vision on Android mobile devices without internet access, usable by farmers? To this end, the most prevalent diseases and pests in the area were analyzed, defining seven priority classes plus one negative control. The classification model used was MobileNetV2 under the CRISP-DM methodology with Transfer Learning and Fine-tuning on a curated dataset. The mobile application was built using the Mobile-D methodology with memory and storage optimizers for low-end devices. The TFLite model achieved an overall accuracy of 96.70 percent and a macro F1-score of 96.65, with an input-range inference latency of 106ms and a file size of 18MB. Usability evaluation with 50 farmers yielded average scores of 4.24, 4.37, and 4.44 for the adapted SUS, TAM, and CSAT instruments, respectively, with a 98 percent satisfaction rate. Technical validation with three agricultural engineers showed a 91.40 percent agreement with experts. These results demonstrate that integrating CRISP-DM and Mobile-D allows for the development of viable artificial intelligence tools for agricultural contexts with infrastructure constraints.
Sistema híbrido basado en modelos de lenguaje de gran escala y análisis estático para la detección automatizada de vulnerabilidades de seguridad en código fuente de aplicaciones web con python y node.js (javascript/typescript).
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software., 2026-07-06)
Terán González, René Alejandro
;
Balarezo López, Julio Enrique
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software.
Static Application Security Testing (SAST) tools are precise but detect few vulnerabilities, whereas large language models (LLMs) widen coverage at the cost of many false positives. This research proposes, implements, and evaluates a hybrid system that combines three SAST configurations (Semgrep, Bandit, and Semgrep JSSec) with five code-specialized LLMs —two commercial (Claude Sonnet 4.6 and Opus 4.7) and three open models run locally on a 6 GB GPU (DeepSeek Coder 16B and 6.7B, Qwen 2.5 Coder 7B)— through an Intelligent Filter that routes each alert and an exploratory mode aimed at business-logic vulnerabilities. The evaluation used a stratified set of 290 Python and JavaScript/TypeScript cases (174 vulnerable and 116 clean), keeping calibration separate from reporting and assessing differences with the exact McNemar test. The best individual LLM (Sonnet 4.6, F1 = 0.643) nearly doubled the F1 of the combined SAST baseline (0.330) without losing precision, and the hybrid system improved F1 over SAST across all three complexity levels (p < 0.001) while containing the false positives of the raw LLM ensemble. The statistical Independence between both paradigms justifies their combination. In an external validation on 7 real CVEs (2024–2026), the system detected 42.9% of the cases and identified a critical SQL injection in Ghost CMS that the three SAST tools missed. Its main contribution is the traceability of every decision and the detection of semantic vulnerabilities using accessible models.
Implementación de un sistema de reportería y visualización de datos con funciones de recomendación y predicción, utilizando técnicas de business intelligence (BI) e inteligencia artificial (IA), en la Dirección de Investigación y Desarrollo (DIDE) de la Universidad Técnica de Ambato (UTA)
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software., 2026-07-06)
Santamaría Márquez, Christopher Paul
;
Santamaría Márquez, Christopher Paul
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software.
This thesis focused on the implementation of a data reporting and visualization system with recommendation and prediction functions, using Business Intelligence and Artificial Intelligence techniques, at the Research and Development Directorate of the Technical University of Ambato. The identified problem was related to the manual management of scientific information, data traceability, and centralization. The research was conducted using an applied, field-based, and non-experimental approach, considering five staff members involved in information management processes as the population. A structured interview with a Likert scale and a complementary semistructured interview were used for the diagnostic process, allowing the identification of eight problems and ten institutional needs. As a result, a web platform based on Scrum + XP was developed, integrating scientific article management, bulk upload, institutional workflow, external integration, ETL processes, an extensible OLTP model, a constellation-type data warehouse, dashboards, and exportable reports. Furthermore, an AI module was implemented with monthly scientific output forecasting, forecasting by faculty and research area, data quality recommendations, editorial recommendations, and AI readiness analysis using weighted indicators. The evaluation showed 80 successful reporting operations, a 100% success rate, an average time of 10.53 seconds, and an estimated reduction of 99.66%. The post-implementation structured interview based on CSUQ received a score of 4.59/5, 100% acceptance, and a Cronbach's alpha of 0.973. It is concluded that the system strengthens information management and supports data-driven decision-making.

Tesis Software

Examinar

Filtros

Ajustes

Ordenar por

resultados por página

Resultados de la búsqueda