One of the popular subsets of ML algorithms is classification algorithms. They are also referred to as supervised learning algorithms. For this approach, we assume that we have a rich dataset of features and events associated with those features. The task of the algorithm is to predict an event given a set of features. The event is referred to as a class variable. For example, consider the following dataset of features related to weather and whether it snowed on a particular day:
Table 1: Sample dataset
|
Temperature (in °F) |
Sky condition |
Wind Speed (in MPH) |
Snowfall |
|
Less than 20 |
Sunny |
30 |
False |
|
20-32 |
Sunny |
6 |
False |
|
32-70 |
Cloudy |
20 |
False |
|
70 and above |
Cloudy |
0 |
False |
|
20-32 |
Cloudy |
10 |
True |
|
32-70 |
Sunny |
15 |
False |
|
Less than 20 |
Cloudy |
8 |
True |
|
32-70 |
Sunny |
7 |
False |
|
20-32... |
