The thyroid-cancer-prediction:ns-rse/evaluate-modelling repository/branch has examples of how to use the TidyModels framework to setup data into training and testing datasets and undertake "machine learning" using a range of models. The example is written in Quarto and rendered to a HTML web-page that is hosted on GitHub and can be viewed here.
Models used...
LASSO regression
ElasticNet
Random Forest
Gradient Boosting
Support Vector Machines (SVM)
Further these are then summarised using a range of classification metrics including plotting Receiver Operating Characteristics and the Area Under the Curve.
Now that work is progressing and @mdp21oe has collected ~1500 cases from historical cases of Thyroid cancer at Sheffield Teaching Hospitals we can progress with adapting this workflow to analysing the data set.
Tasks
[x] Copy r/shf_thy_nod.R from thyroid-cancer-prediction : mdp21oe/sheffield_clean banch. This will serve as the basis for cleaning the dataset and from now on addition and development should be made to this.
The thyroid-cancer-prediction:ns-rse/evaluate-modelling repository/branch has examples of how to use the TidyModels framework to setup data into training and testing datasets and undertake "machine learning" using a range of models. The example is written in Quarto and rendered to a HTML web-page that is hosted on GitHub and can be viewed here.
Models used...
Further these are then summarised using a range of classification metrics including plotting Receiver Operating Characteristics and the Area Under the Curve.
Now that work is progressing and @mdp21oe has collected ~1500 cases from historical cases of Thyroid cancer at Sheffield Teaching Hospitals we can progress with adapting this workflow to analysing the data set.
Tasks
r/shf_thy_nod.R
from thyroid-cancer-prediction :mdp21oe/sheffield_clean
banch. This will serve as the basis for cleaning the dataset and from now on addition and development should be made to this.