HAR dataset
Emerging6papers using it
2020first seen
The HAR dataset is an open-sourced collection of data used to evaluate multi-label classification in human activity recognition, specifically focusing on detecting various movements or actions performed by individuals.
Papers using HAR dataset (6)
- PQS-BFL: A Post-Quantum Secure Blockchain-based Federated Learning FrameworkLabel Leakage in Federated Inertial-based Human Activity RecognitionFed-Focal Loss for imbalanced data classification in Federated LearningFLAME: Federated Learning Across Multi-device EnvironmentsFedSiKD: Clients Similarity and Knowledge Distillation: Addressing
Non-i.i.d. and Constraints in Federated LearningTowards Adaptive Asynchronous Federated Learning for Human Activity
Recognition