HIGHLIGHTS
SUMMARY
Based on these, the authors propose a novel meta-learning based federated personalized cross-domain recommendation model called MFPCDR to solve the user cold-start problem while protecting the user`s private data. The authors propose MFPCDR, a federated personalized cross-domain recommendation model based on meta-learning that employs a federated learning technique to keep user data on the client computer to secure the user`s private data effectively. The authors use meta-networks in the meta-recommendation module to learn personalized transferable embeddings of cold-start users, enabling personalized transfer of user . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.