Figure 1.

An overview of unsupervised and supervised machine learning. Labeled data can be processed in an unsupervised fashion (i.e., without training data) to understand relationships between variables, clustering, etc. Supervised machine learning, on the other hand, requires a training dataset, which can be used to either infer quantitative relationships (regression models) or classify data into training outcomes (classification). Abbreviations: SVM: Support Vector machines; RF: Random Forest. Adapted from Gautam et al. [13].