Recommender-Systems
Recommender-Systems copied to clipboard
nikitaa30
KNN code
Hey nikita, your course was very helpful thank you! i was wondering if you could send me the code (not in the jupiter format) and to ask if it is possible to aply recall and precision metrics in this algorithme Thanks again
the .py version is right there in the same directory by the name 'knn recommender.py'. Also, recall and precision are possible for this algorithm, when your code predicts the rating of the recommended item and compare that with actual or test data.
Thanks a lot! Have a good day
Le mer. 13 mai 2020 à 17:03, Nikita Sharma [email protected] a écrit :
the .py version is right there in the same directory by the name 'knn recommender.py'. Also, recall and precision are possible for this algorithm, when your code predicts the rating of the recommended item and compare that with actual or test data.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/nikitaa30/Recommender-Systems/issues/1#issuecomment-628088001, or unsubscribe https://github.com/notifications/unsubscribe-auth/APSGHJ4643E47SUWMX5JUXTRRLAERANCNFSM4M73OUFA .
Can you give me a quick tip of how can i do it?
Le mer. 13 mai 2020 à 17:03, Nikita Sharma [email protected] a écrit :
the .py version is right there in the same directory by the name 'knn recommender.py'. Also, recall and precision are possible for this algorithm, when your code predicts the rating of the recommended item and compare that with actual or test data.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/nikitaa30/Recommender-Systems/issues/1#issuecomment-628088001, or unsubscribe https://github.com/notifications/unsubscribe-auth/APSGHJ4643E47SUWMX5JUXTRRLAERANCNFSM4M73OUFA .
sure, just found some quick links for you~ If you come across any issue, please let me know https://medium.com/@m_n_malaeb/recall-and-precision-at-k-for-recommender-systems-618483226c54
https://www.researchgate.net/figure/Precision-and-recall-of-recommendations-of-recommender-systems_tbl4_324481236
Amazing! Thank you
Le mer. 13 mai 2020 à 18:04, Nikita Sharma [email protected] a écrit :
sure, just found some quick links for you~ If you come across any issue, please let me know
https://medium.com/@m_n_malaeb/recall-and-precision-at-k-for-recommender-systems-618483226c54
https://www.researchgate.net/figure/Precision-and-recall-of-recommendations-of-recommender-systems_tbl4_324481236
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/nikitaa30/Recommender-Systems/issues/1#issuecomment-628122519, or unsubscribe https://github.com/notifications/unsubscribe-auth/APSGHJ7XBDJ73L6NNKIQDRTRRLHINANCNFSM4M73OUFA .
hello, Sorry for all these questions, but in your knn algorithm we predicted the movies not the ratings, how can we apply recall/ precision then ? Thanks a lot
Le mer. 13 mai 2020 à 17:04, lydia younsi [email protected] a écrit :
Thanks a lot! Have a good day
Le mer. 13 mai 2020 à 17:03, Nikita Sharma [email protected] a écrit :
the .py version is right there in the same directory by the name 'knn recommender.py'. Also, recall and precision are possible for this algorithm, when your code predicts the rating of the recommended item and compare that with actual or test data.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/nikitaa30/Recommender-Systems/issues/1#issuecomment-628088001, or unsubscribe https://github.com/notifications/unsubscribe-auth/APSGHJ4643E47SUWMX5JUXTRRLAERANCNFSM4M73OUFA .
Mine is a content based knn, you should adapt a collaborative approach which outputs ratings, then you can compare those ratings.