-1
$\begingroup$

I have a case where I'm doing text mining over a list of product titles. In particular I want to run a clustering algorithm. But I also have some information about those products that I think can add a lot of information like the category of the product (if the product is a car, a furniture, etc)

When doing the vectorization of the text data I get a big number of text features ~100.000 and I am pretty sure the category information contains much more information than one single text token so the question is what is a reasonable weight for the category information when fitting a model using both the text features and the category information.

$\endgroup$

1 Answer 1

1
$\begingroup$

For obvious reasons we cannot determine the weight for you.

Because there is no mathematical correct weight.

It's entirely up to you to increase the weight of some features you believe to be more important.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.