I have a dataset which contains information about how many times a particular user viewed certain item. So, I don't have rows for all combinations (where the value will be zero ofc because the user has not viewed it), and it would be costly to include this kind of info.

Do I need to include such information if use lightfm, implicit or spotlight library, or they will take care of that in a way that they include them in a calculation (those combinations are important by popular paper http://yifanhu.net/PUB/cf.pdf, because even combinations without interactions have some default value/confidence).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.