Hybrid Support Vector Machine to Preterm Birth Prediction

Santoso, Noviyanti; Wulandari, Sri

doi:10.22146/ijeis.35817

Cited by 6 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The function's basis explains the relationship between the response variable and the predictor. It was suggested that the maximum number of basic functions (BF) is two to four times the number of predictor variables while the maximum number of interactions (MI) is one, two, or three, knowing that more than three results in a very complex model and a minimal distance between knots or minimum observation (MO) of zero, one, two, three, five, and ten [31]. Multivariate Adaptive Regression Spline (MARS) is a non-parametric function that is a complex combination between spline regression and recursive partitioning (RPR) [32].…”

Section: Conceptual Of Multivariate Adaptive Regression Spline (Mars)mentioning

confidence: 99%

ANOVA Decomposition and Importance Variable Process in Multivariate Adaptive Regression Spline Model

Otok¹,

Rumiati²,

Ampulembang³

et al. 2023

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

This article reviews one of the non-parametric functions, namely the MARS (Multivariate Adaptive Regression Spline) method, a complex combination of recursive partitioning and spline regression. The many advantages of the MARS function over other non-parametric regression functions are of interest to researchers. One of them is it can accommodate the additive and interaction model to improve the prediction and interpretation of the data. There are some important things in the MARS method, namely, ANOVA decomposition and Importance Variable. Decomposition ANOVA is a technique in MARS that is useful for grouping basis function based on variables engagement, whether they enter by one variable or interactions with other variables, making it easier to interpret in graphical form. In comparison, the important variables are a technique that can be used to determine which predictor variables most influence the MARS modeling. This study assesses ANOVA decomposition, and the important variables process in MARS modeling based on GCV and MSE criteria. We use the poverty rate modeling data on Java Island to implement the study results. The results show that the MARS model's interpretation of the poverty rate can be better done through ANOVA decomposition. Besides that, based on GCV and MSE criteria, the result also shows that the biggest variable importance in poverty rate modeling on Java Island is the percentage of per capita expenditure for food, while the smallest is the economic growth variable.

show abstract

Section: Conceptual Of Multivariate Adaptive Regression Spline (Mars)mentioning

confidence: 99%

ANOVA Decomposition and Importance Variable Process in Multivariate Adaptive Regression Spline Model

Otok¹,

Rumiati²,

Ampulembang³

et al. 2023

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

show abstract

“…Machine learning has been previously applied to PTB prediction or PTB stratification through many different techniques, such as SVMs 8 , neural networks [9][10][11] and decision trees 12,13 . However, the most commonly used ones are logistic regression and linear regression, employed in the analysis and prediction using many distinct SES facets: poverty 14 , pregnant mother's working conditions 15,16 , general social factors 17,18 and clinical and hereditary factors [19][20][21][22] .…”

Section: Introductionmentioning

confidence: 99%

Self-Organising Maps Applied to the Stratification of Preterm Birth Risk in Brazil with Socioeconomic Data

Lopes,

Silva,

Silva

et al. 2023

Preprint

View full text Add to dashboard Cite

Preterm birth (PTB) poses risks and difficulties for the survival of the newborn baby. Despite considerable progress in research, the causes of PTB still need to be fully understood. PTB risk is believed to be multi-faceted and may be linked to socioeconomic factors. This research aims to categorize the risk of PTB in Brazil based on socioeconomic data. To do this, it will use unsupervised machine learning techniques to identify clusters that show significant differences in PTB risk. By utilizing publicly accessible datasets from the Brazilian Federal Government, a new dataset was created with municipality-level socioeconomic information and a rate of PTB occurrence. This dataset was then processed with unsupervised learning techniques, such as self-organizing maps (SOM) and hierarchical clustering. The research uncovered clusters of municipalities with either high or low levels of PTB. Those with high PTB were mainly composed of municipalities with lower educational attainment, poorer quality of public services (e.g., sanitation and waste disposal), and a smaller white population. The analysis of the regional distribution of clusters revealed that the highest rates of PTB were mainly concentrated in Brazil's North and Northeast regions. The findings suggest that better quality of life and access to public services positively reduce the risk of PTB.

show abstract

“…Várias técnicas de aprendizagem de máquina foram previamente aplicadas ao problema para predic ¸ão ou estratificac ¸ão do risco PTB, incluindo SVMs [8], redes neurais [9]- [11] e árvores de decisão [12], [13]. No entanto, as aplicac ¸ões mais comuns são as técnicas de regressão logística e de regressão linear, empregadas na análise e na predic ¸ão de PTB para diferentes fatores: a pobreza [14], as condic ¸ões de trabalho da gestante [15], [16], fatores sociais em geral [17], [18] e, principalmente, fatores clínicos ou hereditários [19]- [22].…”

Section: Introduc ¸ãOunclassified

Aprendizagem Não-Supervisionada Aplicada à Estratificação de Risco de Nascimentos Prematuros no Brasil com Recorte em Dados Socioeconômicos

Júnior¹,

Barbosa²,

Fernandes³

2021

Anais Do 15. Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

O nascimento prematuro (Preterm birth – PTB) é um fenômeno que traz diversos riscos e desafios à sobrevivência dos recém-nascidos. Apesar de muitos avanços, ainda não foram esclarecidas todas as causas desse fenômeno. Entende-se que o risco ao PTB é multi-fatorial e também pode estar associado a fatores socioeconômicos. Assim, este artigo tem como meta a utilização de técnicas de aprendizagem não-supervisionada para obter um estrato de risco de PTB no Brasil com a partir de dados socioeconômicos. Através de utilização de bancos de dados públicos disponibilizados pelo Governo Federal do Brasil, foi gerado um dataset a nível municipal com informações socioeconômicas e uma taxa de ocorrência de prematuridade. Este dataset foi clusterizado utilizando várias abordagens e técnicas de aprendizagem de máquina não-supervisionada como $k$-means, principal component analysis (PCA) e density-based spatial clustering of applications with noise (DBSCAN) . Após a validação, foram descobertos 4 clusters com ocorrências de PTB muito acima da média nacional, e 3 muito abaixo. Os clusters de alta prematuridade apontaram para municípios de menor nível educacional, de pior qualidade de serviços públicos, como saneamento e coleta de lixo, e de população menos branca. Foi observado ainda o aspecto regional, com os municípios em clusters de maior risco ao PTB estando localizados principalmente nas regiões Norte e Nordeste. Os resultados apontam para uma influência positiva da qualidade de vida e da oferta de serviços públicos na diminuição da ocorrência de nascimentos prematuros.

show abstract

Hybrid Support Vector Machine to Preterm Birth Prediction

Cited by 6 publications

References 10 publications

ANOVA Decomposition and Importance Variable Process in Multivariate Adaptive Regression Spline Model

ANOVA Decomposition and Importance Variable Process in Multivariate Adaptive Regression Spline Model

Self-Organising Maps Applied to the Stratification of Preterm Birth Risk in Brazil with Socioeconomic Data

Aprendizagem Não-Supervisionada Aplicada à Estratificação de Risco de Nascimentos Prematuros no Brasil com Recorte em Dados Socioeconômicos

Contact Info

Product

Resources

About