Open DmitriyPershukov opened 4 months ago
My only idea is that it maybe finds similiar items for items user interacted with and then sorts them by similarity and recommends top k. But im not sure if it is correct.
I know that a lot of time has passed, but...
In this course there is explained user-user and item-item nearest neighbor. You can audit it without payment and giving credit card data :)
Thanks for the link.
On Sun, Jul 21, 2024 at 12:51 AM Jan Sowa @.***> wrote:
I know that a lot of time has passed, but...
In this course there is explained user-user and item-item nearest neighbor. You can audit it without payment and giving credit card data :)
— Reply to this email directly, view it on GitHub https://github.com/benfred/implicit/issues/715#issuecomment-2241277181, or unsubscribe https://github.com/notifications/unsubscribe-auth/ANKOODQVSGL35BNA57KI6XLZNK5SJAVCNFSM6AAAAABHTSKNPOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENBRGI3TOMJYGE . You are receiving this because you authored the thread.Message ID: @.***>
Are there any papers that explain how recommending items to user works initem-item Nearest Neighbour Models?
This page https://benfred.github.io/implicit/api/models/cpu/knn.html# refers to this blog post https://www.benfrederickson.com/distance-metrics/. But as i understand it only explains how you find similiar items.