Transfer Learning for Activity Recognition in Mobile Health
This work addresses the challenge of adapting activity recognition models to new scenarios in mobile health, which is incremental as it builds on existing transfer learning approaches.
The paper tackled the problem of performance degradation in mobile health activity recognition due to differences in sensing platforms and user movement patterns, proposing a transfer learning framework called TransFall that includes a two-tier data transformation, label estimation, and model generation, with validation through analytical and empirical methods.
While activity recognition from inertial sensors holds potential for mobile health, differences in sensing platforms and user movement patterns cause performance degradation. Aiming to address these challenges, we propose a transfer learning framework, TransFall, for sensor-based activity recognition. TransFall's design contains a two-tier data transformation, a label estimation layer, and a model generation layer to recognize activities for the new scenario. We validate TransFall analytically and empirically.