Study of similarity metrics for matching network-based personalised human activity recognition.
Sani, Sadiq; Wiratunga, Nirmalie; Massie, Stewart; Cooper, Kay
Professor Nirmalie Wiratunga email@example.com
Doctor Stewart Massie firstname.lastname@example.org
Senior Research Fellow
Professor Kay Cooper email@example.com
Personalised Human Activity Recognition (HAR) models trained using data from the target user (subject-dependent) have been shown to be superior to non personalised models that are trained on data from a general population (subject-independent). However, from a practical perspective, collecting sufficient training data from end users to create subject-dependent models is not feasible. We have previously introduced an approach based on Matching networks which has proved effective for training personalised HAR models while requiring very little data from the end user. Matching networks perform nearest-neighbour classification by reusing the class label of the most similar instances in a provided support set, which makes them very relevant to case-based reasoning. A key advantage of matching networks is that they use metric learning to produce feature embeddings or representations that maximise classification accuracy, given a chosen similarity metric. However, to the best of our knowledge, no study has been provided into the performance of different similarity metrics for matching networks. In this paper, we present a study of five different similarity metrics: Euclidean, Manhattan, Dot Product, Cosine and Jaccard, for personalised HAR. Our evaluation shows that substantial differences in performance are achieved using different metrics, with Cosine and Jaccard producing the best performance.
|Presentation Conference Type||Conference Paper (unpublished)|
|Start Date||Jul 9, 2018|
|Institution Citation||SANI, S., WIRATUNGA, N., MASSIE, S. and COOPER, K. 2018. Study of similarity metrics for matching network-based personalised human activity recognition. In Minor, M. (ed.) Workshop proceedings for the 26th International conference on case-based reasoning (ICCBR 2018), 9-12 July 2018, Stockholm, Sweden, pages 91-95. Available from: http://iccbr18.com/wp-c...CBR-2018-V3.pdf#page=91|
|Keywords||Human activity recoginition; Matching networks; Data; Case-based reasoning|
SANI 2018 Study of similarity metrics
You might also like
Matching networks for personalised human activity recognition.
Improving human activity recognition with neural translator models.
Zero-shot learning with matching networks for open-ended human activity recognition.
Accuracy of physical activity recognition from a wrist-worn sensor.
Presentation / Conference
Learning deep features for kNN-based human activity recognition.