Representation Learning of Daily Movement Data Using Text Encoders
This work addresses personalized care delivery for people living with Dementia by providing a method to analyze in-home activity data, though it is incremental as it adapts existing text encoders to a specific domain.
The paper tackled the problem of representing daily movement data for remote healthcare monitoring by converting activity recordings into text strings and using a fine-tuned language model to generate embeddings, enabling clustering and identification of deviations for personalized care.
Time-series representation learning is a key area of research for remote healthcare monitoring applications. In this work, we focus on a dataset of recordings of in-home activity from people living with Dementia. We design a representation learning method based on converting activity to text strings that can be encoded using a language model fine-tuned to transform data from the same participants within a $30$-day window to similar embeddings in the vector space. This allows for clustering and vector searching over participants and days, and the identification of activity deviations to aid with personalised delivery of care.