Exceptional Survival Model Mining

Mattos, Juliana Barcellos; Silva, Eraylson G.; Neto, Paulo S. G. de Mattos; Vimieiro, Renato

doi:10.1007/978-3-030-61380-8_21

Cited by 4 publications

(8 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, an important extension of subgroup discovery is the field of exceptional model mining, which allows to apply subgroup discovery algorithms to investigate heterogeneous groups for statistical models. In recent years, exceptional model mining has been developed, for example, for survival analysis (Mattos, Silva, de Mattos Neto, & Vimieiro, 2020), regression models (Duivesteijn, Feelders, & Knobbe, 2012), and mediation models (Lemmerich et al, 2020).…”

Section: Subgroup Discoverymentioning

confidence: 99%

Subgroup Discovery in Structural Equation Models

Kiefer¹,

Lemmerich²,

Langenebrg³

et al. 2021

Preprint

View full text Add to dashboard Cite

Structural equation modeling (SEM) is one of the most popular statistical frameworks in the social and behavioural sciences. Often, detection of groups with distinct sets ofparameters in structural equation models (SEM) are of key importance for appliedresearchers, for example, when investigating differential item functioning for a mentalability test or examining children with exceptional educational trajectories. In this paper, we present a new approach combining subgroup discovery – a well-established toolkit of supervised learning algorithms and techniques from the field of computer science – with structural equation models. We provide an introduction how subgroup discovery can be applied to detect subgroups with exceptional parameter constellations in structural equation models based on user-defined interestingness measures. Furthermore, technical details on the algorithmic components, efficiency, and further computational aspects are presented. Then, our approach is illustrated with two real-world data examples, examining measurement invariance of a mental ability test and investigating interesting subgroups for the mediated relationship between predictors of educational outcomes and the trajectories of math competencies in 5th grade children. The illustrative examples are accompanied bya short introduction in the R package subgroupsem, which is a viable implementation of our approach for applied researchers.

show abstract

Section: Subgroup Discoverymentioning

confidence: 99%

Subgroup Discovery in Structural Equation Models

Kiefer¹,

Lemmerich²,

Langenebrg³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The Exceptional Survival Model Ant-Miner Diverse Search (EsmamDS) is an EMM framework that extends the work presented by Mattos et al [4] to provide a set of more diverse subgroups. Our method tackles redundancy in three different dimensions: description; coverage; and survival model.…”

Section: Esmamds: Exceptional Survival Model Ant-miner -Diverse Searchmentioning

confidence: 99%

“…Alternative approaches were proposed by Mattos et al [4] and Park et al [5]. Both teams tackled the issues raised before by stating the problem as a supervised descriptive pattern mining task.…”

Section: Introductionmentioning

confidence: 99%

“…Both teams tackled the issues raised before by stating the problem as a supervised descriptive pattern mining task. Park et al [5] proposed to tackle it with a subgroup discovery (SD) algorithm, in which the average survival time was considered the target, while Mattos et al [4] proposed an exceptional model mining (EMM) algorithm to find itemset induced subgroups on which a non-parametric survival model was adjusted and compared against a baseline model of the complement.…”

Section: Introductionmentioning

confidence: 99%

“…Approaching the problem as a standard SD task may lead to the loss of interesting patterns due to outliers in the subgroups. The method proposed by Mattos et al [4], on the other hand, is very immature. First, in our preliminary studies, we observed that the patterns returned by the algorithm were extremely redundant.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

EsmamDS: A more diverse exceptional survival model mining approach

Mattos,

Neto,

Vimieiro

2021

Preprint

Self Cite

View full text Add to dashboard Cite

A variety of works in the literature strive to uncover the factors associated with survival behaviour. However, the computational tools to provide such information are global models designed to predict if or when a (survival) event will occur. When approaching the problem of explaining differences in survival behaviour, those approaches rely on (assumptions of) predictive features followed by risk stratification. In other words, they lack the ability to discover new information on factors related to survival. In contrast, we approach such a problem from the perspective of descriptive supervised pattern mining to discover local patterns associated with different survival behaviours. Hence, we introduce the EsmamDS algorithm: an Exceptional Model Mining framework to provide straightforward characterisations of subgroups presenting unusual survival models -given by the Kaplan-Meier estimates. This work builds on the Esmam algorithm to address the problem of pattern redundancy and provide a more informative and diverse characterisation of survival behaviour.

show abstract