Wearable computers provide significant opportunities for sensing and data collection in user's natural environment (NE). However, they require both raw data and annotations to train their respective signal processing algorithms. Collecting these annotations is often burdensome for the users. Our proposed methodology leverages the notion of location from nearable sensors in Internet of Things (IoT) platforms and learns users' patterns of behavior without any prior knowledge. It also requests users for annotations and labels only when the algorithms are unable to automatically annotate the data. We validate our proposed approach in the context of diet monitoring, a significant application that often requires considerable user compliance. Our approach improves eating detection accuracy by 2.4% with requested annotations restricted to 20 per day.
Activity recognition in wearable computing faces two key challenges: i) activity characteristics may be context-dependent and change under different contexts or situations; ii) unknown contexts and activities may occur from time to time, requiring flexibility and adaptability of the algorithm. We develop a context-aware mixture of deep models termed the α-β network coupled with uncertainty quantification (UQ) based upon maximum entropy to enhance human activity recognition performance. We improve accuracy and F score by 10% by identifying high-level contexts in a data-driven way to guide model development. In order to ensure training stability, we have used a clusteringbased pre-training in both public and inhouse datasets, demonstrating improved accuracy through unknown context discovery.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.