5.2.5 Classification
This topic describes the information about the Classification.
Classification is a statistical technique to discover relationships using independent variables to classify a target variable into a number of GROUPS or CLASSES. Mostly used for decision making.
For example: The user needs to predict if a new CASE ID will churn(1) or not (0), given the data of branch, marital status, income and savings.
Table 5-4 Sample Classification
CASE ID | BRANCH | MARITAL STATUS | INCOME | SAVING | CHURN |
---|---|---|---|---|---|
12345 | BRT | Y | 15000 | 6723 | 0 |
12346 | BRT | N | 17500 | 250 | 1 |
12347 | CSR | D | 25000 | 2654 | 1 |
12348 | CSR | N | 16567 | 20 | 0 |
Note:
CASE ID must be unique to identify a row.Parent topic: Frameworks Supported