Can I read Federated Learning for SMS Spam Detection on EtoBox?

Federated Learning for SMS Spam Detection by Øhíñ Hâqùê is a document available to read on EtoBox.

What is Federated Learning for SMS Spam Detection about?

The document presents FedSpam, a federated learning approach for SMS spam prediction that respects user privacy by keeping sensitive SMS data on-device. Utilizing a distilBERT model, the system trains locally on mobile clients and periodically sends model weights to a centralized server for aggregation, achieving high validation accuracy. The methodology aims to create a robust spam classifier that adapts to new spam techniques while ensuring user data confidentiality.

Author

Øhíñ Hâqùê

Language

EN