I have a huge dataset with 3 variables
Company_ID, Area_code, Product_ID each one of them is a categorical variable of levels
Product_ID is the product the
Company_ID is using.Each company can be located in more than one
area_code and can be using more than one product.Lets call this train data.
Given a combination of
Company_ID, Area_code, Product_ID ,
example- comp_025, area_012, p_10 that is not present in train data,i have to predict the probability that a company
Company_ID located in area
Area_Code uses the product
Product_ID. So that i can recommend that product to company.
I am a beginner and having difficulty in coming up with a model. Any input is highly appreciated.