The machine learning techniques can be broadly categorized into supervised

learning $($e$.$g$.,$ classification$)$ and unsupervised learning $($pattern discovery$).$

Supervised learning $($Classification$)$: Objects$/$records$/$data points are classified

into one of a set of pre$-$defined classes. In order to do this, a classification

model is built from a set of data points. The accuracy of the classification of

the model is then evaluated to give some degree of confidence to the result.

Once a reliable classification model has been developed, it is then used to

classify data records whose class outcomes are unknown.

Unsupervised Learning $($Pattern Discovery$)$: Discover underlying patterns and

associations in data when no existing ground truth is known $($i$.$e$.,$ no

observations are tagged$).$ Clustering $($i$.$e$.,$ grouping similar data$)$ can be used

to understand characteristics of certain dis$/$similar data points.