Skip to content

kml : Application of a unsupervised Machine Learning method (clustered longitudinal data analysis using package kml) in R

License

Notifications You must be signed in to change notification settings

audreyyeoCH/Longitudinal_cluster_analysis

Repository files navigation

MSci Biostatistics, University of Zurich, Defence Aug 26 2020.

With special acknowledgement to Dr Roman Flury

Research Aims

  1. What are the covariates that influence the evolution of disease state of Pododermatitis ?
  2. Perform supervised machine learning using package kml (Genolini et al 2010) to cluster longitudinal data to answer (1).
  3. How does scores perform in a robustness checks upon systematic deletion of random individuals ?

Spoiler Alert :

  1. Imputation and missing values
  2. Positive selection bias due to replacement of rabbits
  3. Few time dependant covariates affect scores marginally (sorry!)

Thanks to the 35+ people who attended on zoom + in person. That's a wrap!

About

kml : Application of a unsupervised Machine Learning method (clustered longitudinal data analysis using package kml) in R

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages