Application of Artificial Intelligence techniques for the differentiation of the socioeconomic level

• Autores: Pacori Paucar, Crhistian Ziegler, Mayta Condori, Moises Enrique, Quispe Sanomamani, Luis Fernando, Montana Neyra, Diego Gustavo
• Formato: artículo
• Fecha de Publicación: 2024
• Institución: Universidad La Salle
• Repositorio: Revistas - Universidad La Salle
• Lenguaje: español
• OAI Identifier: oai:ojs.revistas.ulasalle.edu.pe:article/158
• Enlace del recurso: https://revistas.ulasalle.edu.pe/innosoft/article/view/158; https://doi.org/10.48168/innosoft.s15.a158; https://purl.org/42411/s15/a158; https://n2t.net/ark:/42411/s15/a158
• Nivel de acceso: acceso abierto
• Materia: Artificial Intelligence; decision trees; logistic regression; dataset; socioeconomic status; Inteligencia Artificial; árboles de decisión; regresión logística; nivel socioeconómico

In this project, a differentiation is made between people through different parameters such as age, sex, educational level, among others, to try to calculate how much their salary could rise. This problem is important to solve because then a person could predict her future income through the decisions she would make in the present, such as how much education she should receive and when to start working to gain experience. Our procedure to solve this problem has been two statistical analyses, the first linear regression and a decision tree to be able to make a comparison between them, we have tested them using tools such as Colab (Python) and a dataset. Our population for our work was 32,000 records (rows). The results were that through the decision tree there was a precision of 0.88 and an accuracy of 0.82. And with respect to the logistic regression we obtained a precision of 0.80 when for the salary <=50K and 0.72 when the salary is >50K, the accuracy obtained is 0.7912. Concluding that between these two tools we are left with the Decision Tree.