This paper serves as a survey and empirical evaluation of the state-of-the-art in activity recognition methods using accelerometers. The paper is particularly focused on long-term activity recognition in real-world settings. In these environments, data collection is not a trivial matter; thus, there are performance trade-offs between prediction accuracy, which is not the sole system objective, and keeping the maintenance overhead at minimum levels. We examine research that has focused on the selection of activities, the features that are extracted from the accelerometer data, the segmentation of the time-series data, the locations of accelerometers, the selection and configuration trade-offs, the test/retest reliability, and the generalisation performance. Furthermore, we study these questions from an experimental platform and show, somewhat surprisingly, that many disparate experimental configurations yield comparable predictive performance on testing data. Our understanding of these results is that the experimental setup directly and indirectly defines a pathway for context to be delivered to the classifier, and that, in some settings, certain configurations are more optimal than alternatives. We conclude by identifying how the main results of this work can be used in practice, specifically in experimental configurations in challenging experimental conditions.

This paper serves a survey and empirical evaluation of the state-of-the-art in activity 1 recognition methods using accelerometers. We examine research that has focused on the selection 2 of activities, the features that are extracted from the accelerometer data, the segmentation of the 3 time-series data, the locations of accelerometers, the selection and configuration trade-offs, the 4 test/retest reliability, and the generalisation performance. Furthermore, we study these questions 5 from an experimental platform and show, somewhat surprisingly, that many disparate experimental 6 configurations yield comparable predictive performance on testing data. Our understanding of these 7 results is that the experimental setup directly and indirectly defines a pathway for context to be 8 delivered to the classifier, and that, in some settings, certain configurations are more optimal than 9 alternatives. We conclude by identifying how the main results of this work can be used in practice, 10 specifically in experimental configurations in challenging experimental conditions.

No abstract

Deep clustering has increasingly been demonstrating superiority over conventional shallow clustering algorithms. Deep clustering algorithms usually combine representation learning with deep neural networks to achieve this performance, typically optimizing a clustering and non-clustering loss. In such cases, an autoencoder is typically connected with a clustering network, and the final clustering is jointly learned by both the autoencoder and clustering network. Instead, we propose to learn an autoencoded embedding and then search this further for the underlying manifold. For simplicity, we then cluster this with a shallow clustering algorithm, rather than a deeper network. We study a number of local and global manifold learning methods on both the raw data and autoencoded embedding, concluding that UMAP in our framework is able to find the best clusterable manifold of the embedding. This suggests that local manifold learning on an autoencoded embedding is effective for discovering higher quality clusters. We quantitatively show across a range of image and timeseries datasets that our method has competitive performance against the latest deep clustering algorithms, including out-performing current state-of-the-art on several. We postulate that these results show a promising research direction for deep clustering. The code can be found at https://github.com/rymc/n2d.

No abstract