Software
Department of Medical Statistics
R-package glmmml
Description: Combine random effects modeling with machine learning (ML).
Iterate between fitting a (fixed-effects only) ML model and a generalized linear model to estimate the random effects. In each case use the estimations from the other step as offset to the data.
The idea follows Ngufor et al., 2019, Journal of Biomedical Informatics (doi:10.1016/j.jbi.2018.09.001) the implementation draws from glmertree.
R-package hi2
Description: Molecular diagnosis or prediction of clinical treatment outcome based on high-throughput genomics data is a modern application of machine learning techniques for clinical problems. In practice, clinical parameters, such as patient health status or toxic reaction to therapy, are often measured on an ordinal scale (e.g. good, fair, poor).
Commonly, the prediction of ordinal end-points is treated as a multi-class classification problem, disregarding the ordering information contained in the response. hierarchical twoing (hi2) combines the power of well-understood binary classification with ordinal response prediction.
Reference: Andreas Leha, Klaus Jung, and Tim Beißbarth (2013). "Utilization of ordinal response structures in classification with high-dimensional expression data". In: German Conference on Bioinformatics 2013. OpenAccess Series in Informatics (OASIcs). Dagstuhl, Germany, pp. 90–100. https://doi.org/10.4230/OASIcs.GCB.2013.90
R-package cvdf
Description: Quickly generate cross validation (CV) structures. Make CVs easier manageable. The structure is initiated in a data.frame (DF) which facilitates dplyr based code. Y times repeated X-fold CVs are supported in stratified and unstratified form. Provides functionality to generate the CV structure, to pre-partition the data using a given/generated CV structure and to impute data within the folds.
This could also be of interest