Bibliographic citations
Lazo, U., (2023). Aplicación de un algoritmo no supervisado para clasificar extranjeros según riesgo de inmigración irregular en la Superintendencia Nacional de Migraciones [Universidad Nacional de San Martín. Fondo Editorial]. http://hdl.handle.net/11458/5170
Lazo, U., Aplicación de un algoritmo no supervisado para clasificar extranjeros según riesgo de inmigración irregular en la Superintendencia Nacional de Migraciones []. PE: Universidad Nacional de San Martín. Fondo Editorial; 2023. http://hdl.handle.net/11458/5170
@mastersthesis{renati/1054027,
title = "Aplicación de un algoritmo no supervisado para clasificar extranjeros según riesgo de inmigración irregular en la Superintendencia Nacional de Migraciones",
author = "Lazo Bartra, Ulises",
publisher = "Universidad Nacional de San Martín. Fondo Editorial",
year = "2023"
}
Application of an unsupervised algorithm to classify foreigners according to risk of irregular immigration in the National Superintendence of Migration. Migration is a social phenomenon that affects the structure and distribution of the population and is motivated by the search for better opportunities and living conditions. However, irregular migration poses challenges for the receiving countries, since it involves the entry of people without proper documentation and can generate insecurities in terms of national security and border control. In this context, this study aimed to evaluate the application of an unsupervised algorithm to classify foreigners according to the risk of irregular immigration in the National Superintendence of Migration. For this purpose, the DBSCAN (Density-Based Spatial Clustering of Noisy Applications) algorithm was used. The methodology of the study involved the construction of a dataset with data on foreigners reported by the National Superintendence of Migration, the application of the DBSCAN algorithm to classify foreigners into different clusters according to their level of risk of irregular immigration, and the determination of the Silhouette coefficient as a measure of the quality of the classification. The results showed that the DBSCAN unsupervised learning algorithm was able to classify foreigners into four clusters representing the levels of irregular immigration risk: high, medium high, medium low and low. The Silhouette coefficient obtained was 0.534, indicating a significant and consistent classification. In conclusion, this study demonstrated that the application of DBSCAN as an unsupervised algorithm is an effective strategy for the classification of foreigners according to the risk of irregular immigration in the National Superintendency of Migration. These results have important implications for informed decision making by migration inspectors, contributing to the reduction of irregular immigration and the maintenance of appropriate immigration status in the country.
This item is licensed under a Creative Commons License