RESUMO
Data mining is employed to extract useful information and to detect patterns from often large data sets, closely related to knowledge discovery in databases and data science. In this investigation, we formulate models based on machine learning algorithms to extract relevant information predicting student retention at various levels, using higher education data and specifying the relevant variables involved in the modeling. Then, we utilize this information to help the process of knowledge discovery. We predict student retention at each of three levels during their first, second, and third years of study, obtaining models with an accuracy that exceeds 80% in all scenarios. These models allow us to adequately predict the level when dropout occurs. Among the machine learning algorithms used in this work are: decision trees, k-nearest neighbors, logistic regression, naive Bayes, random forest, and support vector machines, of which the random forest technique performs the best. We detect that secondary educational score and the community poverty index are important predictive variables, which have not been previously reported in educational studies of this type. The dropout assessment at various levels reported here is valid for higher education institutions around the world with similar conditions to the Chilean case, where dropout rates affect the efficiency of such institutions. Having the ability to predict dropout based on student's data enables these institutions to take preventative measures, avoiding the dropouts. In the case study, balancing the majority and minority classes improves the performance of the algorithms.
RESUMO
El artículo propone analizar y constatar la vigencia dealgunos de los postulados del economista estadounidense, Milton Friedman, y su relación con la política pública, en específico para el caso de Chile, y las políticas en materia de envejecimiento. Desdeuna aproximación sociohistórica y biopolítica al tratamiento que la vejez ha tenido a nivel público, se analizan los postulados de Friedman, y como estos promueven la construcción de un sujeto activo-exitoso, como único responsable de su devenir. Finalmente se comparten reflexiones que podrían contribuir a re-elaborar y tensionar las propuestas de envejecimiento positivo-activopara Chile, y Latinoamérica.
The article proposes to analyze and verify the validity of some of the postulates of the American economist, Milton Friedman, and their relationship with public policy, specifically in the case of Chile, and aging policies. From a sociohistorical and biopolitical approach to the treatment that old age has had at the public level, Friedman's postulates are analyzed, and how they promote the construction of an active-successful subject, as the only person responsible for their future. Finally, reflections are shared that could contribute to re-elaborate and discuss the proposals of positive-active aging for Chile, and Latin America.
O artigo propõe analisar e verificar a validade de alguns dos postulados do economista americano Milton Friedman e sua relação com políticas públicas, especificamente no caso do Chile, e políticas de envelhecimento.De uma abordagem sócio-histórica e biopolítica ao tratamento que a velhice teve no nível público, são analisados os postulados de Friedman e como eles promovem a construção de um sujeito ativo e bem-sucedido, como aúnica pessoa responsável por seu futuro. Por fim, são compartilhadas reflexões que poderiam contribuir para reelaborar e enfatizar aspropostas de envelhecimento positivo e ativopara o Chile e a América Latina.
Assuntos
Humanos , Masculino , Feminino , Idoso , Idoso de 80 Anos ou mais , Política Pública/economia , Envelhecimento Saudável , Chile , Política de Saúde/legislação & jurisprudênciaRESUMO
ABSTRACT Background: The burden of obstructive sleep apnea (OSA) in rural settings is unknown. In these regions, devices needed for OSA diagnosis are not available, and mass screening with field instruments may be complicated due to cross-cultural factors and illiteracy. The association between the Friedman's tongue position (FTP) and OSA has been assessed in people from different ethnic groups but not in Amerindians. Objective: We aimed to assess whether a FTP type IV is associated with OSA severity and with the apnea-hypopnea index (AHI) in community-dwelling older adults of Amerindian ancestry living in rural Ecuador. Methods: A total of 201 Atahualpa residents aged ≥60 years, who underwent tongue position assessment, brain MRI, and polysomnography were included. After adjusting for relevant confounders, ordinal logistic regression models were fitted to assess the association between the presence of a FTP type IV and OSA categories (none, mild, and moderate-to-severe), and generalized linear models with a Gaussian link were fitted to assess the association between the presence of a FTP type IV and the continuous AHI. Results: A FTP type IV was identified in 153 (76%) individuals, the mean AHI per hour was 11.9 ± 12.4, and 49 (24%) individuals had moderate-to-severe OSA, 88 (44%) had mild OSA, and the remaining 64 (32%) had no OSA. Fully-adjusted generalized linear models showed no independent association between the investigated exposure and the AHI (β: 0.09; 95% C.I.: -1.56 - 1.76; p=0.909). Likewise, ordinal logistic regression models showed no independent association between the investigated exposure and categories of OSA (β: 0.42; 95% C.I.: -0.47 - 1.31; p=0.357). Conclusion: A FTP type IV is not associated with the AHI or the severity of OSA in this population of Amerindians. This lack of association could be related to phenotypic characteristics of people from this ethnic group (mostly their elliptic hard palate).
RESUMEN Antecedentes: Se desconoce la prevalencia de la apnea obstructiva del sueño (AOS) en entornos rurales. En esas regiones, los equipos necesarios para el diagnóstico de AOS no están disponibles, y la detección de AOS con instrumentos de campo puede ser complicada debido a factores interculturales y analfabetismo. La asociación entre la posición de la lengua de Friedman (FTP) y AOS se ha evaluado en personas de diferentes grupos étnicos, pero no en Amerindios. Objetivo: Evaluar si al tipo IV de FTP está asociado con la severidad de la AOS y con el índice de apnea-hipopnea (IAH) en adultos mayores que viven en una comunidad de ascendencia Amerindia en zonas rurales de Ecuador. Métodos: Se incluyeron 201 residentes de Atahualpa de edad ≥60 años, que fueron sometidos a evaluación de la posición de la lengua, resonancia magnética cerebral y polisomnografía. Después de ajustar por factores de confusión relevantes, modelos de regresión logística ordinal evaluaron la probable asociación entre la presencia de un FTP tipo IV y las categorías AOS (ninguna, leve y moderada a grave), y se ajustaron modelos lineales generalizados con un enlace gaussiano para evaluar la asociación entre la presencia de un FTP tipo IV y el IAH continuo. Resultados: Se identificó un tipo IV de FTP en 153 (76%) individuos, el IAH promedio por hora fue de 11.9 ± 12.4 y 49 (24%) individuos tenían AOS de moderada a grave, 88 (44%) tenían AOS leve. y los 64 restantes (32%) no tenían AOS. Los modelos lineales generalizados, ajustados por confusores, no mostraron una asociación independiente entre la exposición investigada y el IAH (β: 0.09; 95% C.I.: -1.56 - 1.76; p = 0.909). Del mismo modo, los modelos de regresión logística ordinal no mostraron una asociación independiente entre la exposición investigada y las categorías de AOS (β: 0,42; 95% C.I.: -0,47 - 1.31; p = 0,357). Conclusión: El tipo IV de FTP no está asociado con el IAH o la gravedad de la AOS en esta población de Amerindios. Esta falta de asociación podría estar relacionada con las características fenotípicas de las personas de este grupo étnico (principalmente su paladar óseo de tipo elíptico).
RESUMO
Graph derivative indices (GDIs) have recently been defined over N-atoms (N = 2, 3 and 4) simultaneously, which are based on the concept of derivatives in discrete mathematics (finite difference), metaphorical to the derivative concept in classical mathematical analysis. These molecular descriptors (MDs) codify topo-chemical and topo-structural information based on the concept of the derivative of a molecular graph with respect to a given event (S) over duplex, triplex and quadruplex relations of atoms (vertices). These GDIs have been successfully applied in the description of physicochemical properties like reactivity, solubility and chemical shift, among others, and in several comparative quantitative structure activity/property relationship (QSAR/QSPR) studies. Although satisfactory results have been obtained in previous modelling studies with the aforementioned indices, it is necessary to develop new, more rigorous analysis to assess the true predictive performance of the novel structure codification. So, in the present paper, an assessment and statistical validation of the performance of these novel approaches in QSAR studies are executed, as well as a comparison with those of other QSAR procedures reported in the literature. To achieve the main aim of this research, QSARs were developed on eight chemical datasets widely used as benchmarks in the evaluation/validation of several QSAR methods and/or many different MDs (fundamentally 3D MDs). Three to seven variable QSAR models were built for each chemical dataset, according to the original dissection into training/test sets. The models were developed by using multiple linear regression (MLR) coupled with a genetic algorithm as the feature wrapper selection technique in the MobyDigs software. Each family of GDIs (for duplex, triplex and quadruplex) behaves similarly in all modelling, although there were some exceptions. However, when all families were used in combination, the results achieved were quantitatively higher than those reported by other authors in similar experiments. Comparisons with respect to external correlation coefficients (q2ext) revealed that the models based on GDIs possess superior predictive ability in seven of the eight datasets analysed, outperforming methodologies based on similar or more complex techniques and confirming the good predictive power of the obtained models. For the q2ext values, the non-parametric comparison revealed significantly different results to those reported so far, which demonstrated that the models based on DIVATI's indices presented the best global performance and yielded significantly better predictions than the 12 0-3D QSAR procedures used in the comparison. Therefore, GDIs are suitable for structure codification of the molecules and constitute a good alternative to build QSARs for the prediction of physicochemical, biological and environmental endpoints.