all the customers who spend more than 800 rupees in a single purchase are categor-
ized as good customers (this is a hypothetical example or analysis).
From the next time onwards, the marketing manager can plot the new customers
on this graph and based on which side they fall, predicting whether the customer is
likely to be good or bad.
Note that classification need not always be binary (yes or no, male or female, good
or bad, and so on). Any number of classifications is feasible (poor, below average,
average, above average, or good) based on the problem definition. Also, note that
in regression what you are finding is a continuous value and in classification, it takes
only a few values.
This analytical procedure is referred to as supervised learning as the data on which
we operate is known to us and the expectation on what needs to be analyzed from
the data is defined.
Forecasting or prediction or regression
Forecasting or prediction is all about the way things would happen in future. This in-
formation can be derived from past experience or knowledge. In this case, we can
have little data and through regression we end up defining the future. Forecasting
and prediction results are usually presented along with the degree of uncertainty.