I am working on the KDD dataset given in this link.
The dataset is related to a typical recommendation systems dataset. So you find an item and information about the item. One of the information given about the Item is its category. Item-Category is a string
a.b.c.d, where the character delimits the categories in the hierarchy
., ordered in a top-down fashion (i.e., category
a is a parent category of
b, and category
b is a parent category of
c, and so on.
I'm not sure how to use this information correctly in my feature engineering. For example, the simplest information I can derive is for each Item I can estimate the topmost category to which it belongs. However, to go beyond this and use the subcategory information, how can I model this feature for a linear regression?