Publication: Chatbot para la recuperación de información personal
Loading...
Official URL
Full text at PDC
Publication Date
2023-01-24
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citations
Exportar
Abstract
Este trabajo se centra en el desarrollo de un Chatbot que recopile y almacene los recuerdos de una persona con demencia. El objetivo es facilitar la tarea de los terapeutas con ese paso previo a la construcción de la historia de vida de un paciente. Se trata de una herramienta de apoyo de cara a las terapias basadas en reminiscencia.
El chatbot se ha programado en base a tres componentes principales. El primero es un analizador de sentimientos que detecta si un texto tiene connotaciones negativas o positivas. En segundo lugar se encuentra el clasificador de recuerdos en etapas de vida que indica si el texto se corresponde con un recuerdo de la infancia, de la juventud, de la etapa adulta o de la vejez. El último componente es un módulo que elige cuál es la mejor siguiente pregunta a formular, la que más relación tenga con la respuesta que haya dado el usuario.
Para que la interacción con el Chatbot sea atractiva, visual y fácil, se ha creado una aplicación web complementaria dirigida tanto a terapeutas como a pacientes. Aunque el foco principal de la aplicación sea el bot conversacional, también se ofrecen otras funcionalidades como la posibilidad de revisar los recuerdos ya almacenados o la de crear nuevas terapias.
El código desarrollado en el proyecto puede encontrarse en https://github.com/NILGroup/TFG-2122-Chatbot.
This project focuses on the development of a Chatbot that collects and stores the memories of people who suffer dementia. The main goal is to facilitate the therapist's job by helping with the steps prior to shapping a patient's life story. It's a support tool for reminiscence-based therapies. The Chatbot has been programmed considering three main components. The first one is a sentiment analyser that detects whether the text contains negative or positive connotations. Secondly, we have a memory classifier that indicates whether the text corresponds to a memory from the childhood, youth, adulthood or old age. The last component is a module that chooses the next best question to be asked, the one that is most related to the answer given by the user. To make the interaction with the Chatbot attractive, visual and easy, a complementary web application has been created for both therapists and patients. Although the main focus of the app is the conversational bot, other functionalities are also offered, such as the possibility of reviewing the already stored memories or creating new therapies. The code developed throughout the project can be found at https://github.com/NILGroup/TFG-2122-Chatbot.
This project focuses on the development of a Chatbot that collects and stores the memories of people who suffer dementia. The main goal is to facilitate the therapist's job by helping with the steps prior to shapping a patient's life story. It's a support tool for reminiscence-based therapies. The Chatbot has been programmed considering three main components. The first one is a sentiment analyser that detects whether the text contains negative or positive connotations. Secondly, we have a memory classifier that indicates whether the text corresponds to a memory from the childhood, youth, adulthood or old age. The last component is a module that chooses the next best question to be asked, the one that is most related to the answer given by the user. To make the interaction with the Chatbot attractive, visual and easy, a complementary web application has been created for both therapists and patients. Although the main focus of the app is the conversational bot, other functionalities are also offered, such as the possibility of reviewing the already stored memories or creating new therapies. The code developed throughout the project can be found at https://github.com/NILGroup/TFG-2122-Chatbot.
Description
Trabajo de Fin de Grado en Ingeniería Informática, Facultad de Informática UCM, Departamento de Ingeniería del Software e Inteligencia Artificial, Curso 2022/2023.