I'm working with http://archive.ics.uci.edu/ml/datasets/Bank+Marketing# dataset in order to create a model. We're going to use it in a presentation to introduce people our new data science environment. In the dataset, pdays column represents how many days passed over after previous campaign to customer. When there is no previous campaign for a specific customer then pdays is set to 999. Is it true to use it as so? Maybe I can create a new flag represents whether there is a previous campaign for a specific user or maybe I can set it simply 0 if there is no previous campaign. For example, if I'm using logistic regression then linear combination of features are used and for customers whose pdays is 0(no previous campaign) then simply pdays column has no effect. What would the correct treatment?