DynImp: Dynamic Imputation for Wearable Sensing Data Through Sensory and Temporal Relatedness
This work addresses data quality issues in wearable sensing for applications like health monitoring, though it is incremental as it builds on existing imputation and LSTM-based methods.
The paper tackled the problem of irregularly sampled or missing data in wearable sensing by proposing DynImp, a model that uses sensory and temporal relatedness to reconstruct data under extreme missingness (>50% missing rate), achieving improved performance in activity recognition tasks.
In wearable sensing applications, data is inevitable to be irregularly sampled or partially missing, which pose challenges for any downstream application. An unique aspect of wearable data is that it is time-series data and each channel can be correlated to another one, such as x, y, z axis of accelerometer. We argue that traditional methods have rarely made use of both times-series dynamics of the data as well as the relatedness of the features from different sensors. We propose a model, termed as DynImp, to handle different time point's missingness with nearest neighbors along feature axis and then feeding the data into a LSTM-based denoising autoencoder which can reconstruct missingness along the time axis. We experiment the model on the extreme missingness scenario ($>50\%$ missing rate) which has not been widely tested in wearable data. Our experiments on activity recognition show that the method can exploit the multi-modality features from related sensors and also learn from history time-series dynamics to reconstruct the data under extreme missingness.