I am working for a client who is in e-commerce. They have thousands of offerings and each has its own set of tags. Now, they want to find similar designs dynamically based on the item selected by the customer.
So, I would like to build a syntax based similarity measure. i.e. I give tags against each offering along with a target offering; the system gives related offerings to the target offering in an order (Rank-wise).
Is there any similar library available in R or other statistical tools? If not, any suggestion on how to develop the algorithm or any reference materials you can point to?