Proceedings of the 7th International Conference on Advanced Research in Education
Year: 2024
DOI:
[PDF]
Federated Learning Based Heart Disease Prediction in Smart Cities
Srinivasachar, Rajashekara R, Sughosh Athreya K N, Thilak T
ABSTRACT:
Federated learning, a decentralized model training paradigm, is implemented using the Flower framework in this research. The study focuses on predicting heart disease collaboratively across multiple clients while preserving data privacy. A custom neural network model is trained on clients, and the federated learning strategy of Flower facilitates model updates aggregation. The experiments showcase the effectiveness of collaborative model training, emphasizing data privacy preservation and reduced communication overhead. The results contribute to the growing field of federated learning applications and highlight the unique features of the Flower framework.
keywords: Federated Learning, Heart Disease, Healthcare, Decentralized data