Conditional probability occurs when there is a possibility that an event occurs after another event has happened previously, Bayes’s theorem is a proposition that is used to calculate the probability with several conditions of an event.
For this exercise we have a call center data set of complaints and we want to calculate probabilities according to the complaints made about animals (bee threat, dangerous animal)
The question is: What is the probability that the animal report will be HIGH on a Monday early morning?
For that we calculate the HIGH and NOT HIGH animal report and all the conditions.
- Animal complaints High and not High from Monday to Sunday
- Animal complaints High and not High by schedules from early morning to night
After the exercise we can obtain the probabilities per day and time shift
We make a chart and see the trend lines of the probabilities.
These help us to have a pattern of possible events, which we can use for prediction or to include it as values in linear, logistic or neural network regressions.
If we compare it with the same data but in a relative frequency table, we can see that the higher number probabilities do happen.
