I am working in anaconda/python and I have a datase which contains userId ,itemId and 2 more attribute and timestamp for purchase.

UserId     ItemId   attr1   attr2   timestamp
u1             i2   1.56    2.15     Fri Aug 10 16:40:07 +0000 2018
u1             i1   2.56    2.45     Sat May 12 18:40:38 +0000 2018
u2             i1   3.66    4.5      Wed Apr 18 16:07:03 +0000 2018

normally CF uses rating attribute to find correlation between users using pearson,cosine and other correlation algorithm. But how can I use attr1,attr2 and timestamp to find correlation instead of rating attribute which I don't have?.


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.