A label, or target variable, is the output a supervised model is trained to predict. Each historical training example pairs a known label with input attributes, also called independent variables. The model learns patterns from those labeled examples to predict outputs for unseen data. section 1.5.1
The target determines the prediction task: a continuous value such as revenue calls for regression, while a category such as fraudulent or genuine calls for classification. Unsupervised learning instead looks for patterns in input data without referring to a target variable. section 1.5.1 section 1.5.2
Table 1.3 Example of a structured dataset for a telecom operator showing multiple data attributes
|
ID
|
Revenue ($)
|
Duration of service (years)
|
Avg. cost
|
Monthly usage (days)
|
Churned (Y/N)
|
|---|---|---|---|---|---|
|
1001
|
100
|
1.1
|
0.10
|
10
|
Y
|
|
1002
|
200
|
4.1
|
0.09
|
25
|
N
|
|
1003
|
300
|
5.2
|
0.05
|
28
|
N
|
|
1004
|
200
|
0.9
|
0.25
|
11
|
Y
|
|
1005
|
100
|
0.5
|
0.45
|
12
|
Y
|
In the telecom dataset, “Churned (Y/N)” is the label to predict; revenue, service duration, average cost, and monthly usage are input attributes. section 1.5.1
