Regression framework
When you consider the different types of fields in ServiceNow, you might notice that not all fields lend themselves to a classification approach. In addition to assigning categorical values, you may also want to predict numeric field values such as durations, sizes, effort levels, or other measures that vary along some numeric scale. Regression solutions have some distinctive characteristics, which make them very useful as an additional tool for automating processes. To understand these benefits, it is necessary to think about the types of data that we would be predicting and how a person would estimate the number based on inputs. Let’s use the example of estimating the hours that should be estimated for a user story based on some given input text and an assigned team.
The first thing to think about in this situation is whether we should hope for a precisely correct answer each time. If you estimated 30 hours and it turned out to be 28 or 32, you’...
