Modelo predictivo de desnutrición infantil en el Ecuador: distrito zona 5

In Ecuador, child malnutrition has become a public health challenge, impacting children's development. This incidence affects not only the physical aspect but also the cognitive, emotional, and social dimensions, posing a disadvantage throughout their life cycle. The implementation of social programs and the availability of tools to combat the rise, evaluate factors, and predict child malnutrition has been a challenge for the country's authorities. In response to this situation, this work aims to provide a model that helps predict child malnutrition, focusing on a specific area of Ecuador, zone 5 (Santa Elena, Guayas, Los Ríos, Galápagos). To achieve this, a database from the Ministry of Public Health, including child patient records from 2021 and 2022, was used. Data collection, unification, cleaning, treatment of missing values, and normalization of variables are among the techniques used in the study. Machine learning algorithms such as logistic regression, Random Forest, K-Nearest Neighbors, classification tree, and XGBoost were used to evaluate the model. The results indicate that XGBoost has the highest accuracy in predicting child malnutrition. Key indicators such as age, weight, height, and body mass index were identified through data analysis; these are essential for assessing children's nutritional status. The predictive model has proven to be a useful tool for identifying early malnutrition, which would help in implementing more effective preventive and therapeutic interventions

• Päätekijä: Carrión González, Angélica Neomí (author)
• Aineistotyyppi: masterThesis
• Kieli: spa
• Julkaistu: 2024
• Aiheet: Desnutrición Infantil; Modelo predictivo; Aprendizaje automático; Salud pública
• Linkit: https://repositorio.uteq.edu.ec/handle/43000/7839