r-vtype

Estimates the type of variables in non-quality controlled data. The prediction is based on a random forest model, trained on over 5000 medical variables with accuracy of 99%. The accuracy can hardy depend on type and coding style of data.