Nominal data is basically data which can be neither ordered nor measured it's merely different categories, for example, color - all are similar they don't have any hierarchical order.
There are many ways to deal with nominal data for predictive modeling - which approach to choose is based on the scenario and the kind of predictive modeling/visualization you would like to do.
Various approaches can be found on the following link: https://bit.ly/2mHCt6m