Métodos de recuperación para sistema de razonamiento basado en casos de pacientes con dislalia.
No hay miniatura disponible
Fecha
2021
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad de Guayaquil. Facultad de Ciencias Matemáticas y Físicas. Carrera de Ingeniería en Sistemas Computacionales.
Resumen
El presente trabajo de titulación perteneciente a un proyecto FCI está orientado en la exploración y estudio de los mejores métodos para la recuperación de casos de pacientes con dislalia, en la búsqueda de agilizar la asignación de tratamientos a dichos pacientes. En la actualidad son pocos los centros que tratan este trastorno y en donde se puede observar que se denota una gran falta de soluciones informáticas que brinden ayuda en cuanto cómo se maneja el proceso de diagnóstico y tratamiento. Por lo tanto, se ve conveniente definir un método de recuperación mediante la adaptación de un algoritmo que determine óptimamente el nivel de similitud apropiado y acertado. Para lograr que el diagnóstico presentado sea el adecuado se analizó una base de casos previamente realizada en un trabajo de titulación perteneciente a la misma línea de investigación, en conjunto con una comparación de varios algoritmos de recuperación encontrados durante la realización de la investigación. Para el desarrollo de este proyecto se consultaron diversas fuentes seguras, los temas más necesarios con respecto a la información requerida sobre la condición del trastorno del lenguaje, a su vez se ha trabajado con las metodologías de investigación científica para la consulta de conceptos y la metodología empírica para el desarrollo del algoritmo, mismo que ha sido construido en el lenguaje de programación Python. Como conclusión se tiene que el uso del algoritmo de regresión logística ha resultado ser el más apropiado en comparación con el de k-vecinos más cercanos (k-NN) demostrando mayor rango de precisión y exactitud durante todo el proceso al obtener predicciones acertadas basada en los casos más parecidos.
The present degree work belongs to a FCI project oriented in the exploration and study of the best methods for the recovery of cases of patients with dyslalia, in the search for the fastest assign for treatments for these patients. At present there are very a few centers for this disorder and where it can be observer that there are a deficient informatic solution to help the process of diagnosis and treatment. Therefore, it seems convenient to define a recovery method through the adaptation of an algorithm that optimized determine an appropriate and correct level of similarity. To ensure the presented diagnosis is accurate it was analyzed a base of previous cases by a previous work of degree in the same line of investigation, altogether with a comparison of various algorithms of recovery founded during the investigation. For the development of this project various safety sources were consulted, the most necessary subjects of information regarding this language disorder were required, at the same time methods of scientific investigation were worked for the consult of concepts and empiric methodology for the development of the algorithm, same that was rebuild in Python programming language. In conclusion there is that the use of the linguistic regression algorithm turns to be the most appropriated in comparison with the k-nearest neighbors (k-NN) demonstrating highest range of precision and exactitude during all the process to obtain accurate predictions in base with most similar cases.
The present degree work belongs to a FCI project oriented in the exploration and study of the best methods for the recovery of cases of patients with dyslalia, in the search for the fastest assign for treatments for these patients. At present there are very a few centers for this disorder and where it can be observer that there are a deficient informatic solution to help the process of diagnosis and treatment. Therefore, it seems convenient to define a recovery method through the adaptation of an algorithm that optimized determine an appropriate and correct level of similarity. To ensure the presented diagnosis is accurate it was analyzed a base of previous cases by a previous work of degree in the same line of investigation, altogether with a comparison of various algorithms of recovery founded during the investigation. For the development of this project various safety sources were consulted, the most necessary subjects of information regarding this language disorder were required, at the same time methods of scientific investigation were worked for the consult of concepts and empiric methodology for the development of the algorithm, same that was rebuild in Python programming language. In conclusion there is that the use of the linguistic regression algorithm turns to be the most appropriated in comparison with the k-nearest neighbors (k-NN) demonstrating highest range of precision and exactitude during all the process to obtain accurate predictions in base with most similar cases.
Descripción
PDF
Palabras clave
Dislalia, Métodos de Recuperación, Algoritmo, Base de casos, Trastorno del Lenguaje, Python, Dyslalia, Case Basis, Recovery Methods, Algorithm, Language Disorder