Multivariate Human Activity Segmentation: Systematic Benchmark with ClaSP
Human activity recognition (HAR) systems extract activities from observational data, such as sensor measurements from mobile devices, to provide for instance medical, fitness, or security information. A crucial initial step in these data analysis workflows is segmenting continuous numerical measurements into variable-sized segments that correspond to single activities. Human activity segmentation (HAS) enables downstream classification algorithms to label entire activities. Unfortunately, current time series segmentation (TSS) algorithms exhibit limited performance on multivariate sensor data due to complex temporal dynamics and irrelevant dimensions. This limits their applicability in HAR workflows. In this review, we provide a systematic benchmark of dimensionality reduction, model aggregation, and change point selection applied to the ClaSP TSS algorithm for real-world, multidimensional mobile sensing data. We evaluated the accuracy of the techniques in an experimental study using 250 data sets from the HAS challenge at ECML/PKDD and AALTD 2023. Our findings indicate that extending ClaSP for multivariate data, by aggregating internal representations, yields better results compared to reducing data dimensionality or selecting change points (CPs) from different channels. We report a new state of the art with 73% average accuracy on the challenge benchmark.
Code (1)
Tasks
Activity RecognitionChange Point DetectionDimensionality ReductionHuman Activity RecognitionSegmentationSemantic SegmentationTime SeriesTime Series AnalysisSimilar Papers 제목 키워드 기반
Human Activity Segmentation Challenge @ ECML/PKDD’23
Time series segmentation (TSS) is a research problem that focuses on dividing long multivariate sensor data into smaller, homogeneous subsequences. This task is critical for various real-world data analysis applications,…
Activity RecognitionChange Point DetectionHuman Activity RecognitionSegmentation+3Joint segmentation of multivariate time series with hidden process regression for human activity recognition
The problem of human activity recognition is central for understanding and predicting the human behavior, in particular in a prospective of assistive services to humans, such as health monitoring, well being, security, e…
Activity RecognitionHuman Activity RecognitionregressionTime Series+1Are uGLAD? Time will tell!
We frequently encounter multiple series that are temporally correlated in our surroundings, such as EEG data to examine alterations in brain activity or sensors to monitor body movements. Segmentation of multivariate tim…
EEGElectroencephalogram (EEG)SegmentationTime SeriesA Multi-Task Deep Learning Approach for Sensor-based Human Activity Recognition and Segmentation
Sensor-based human activity segmentation and recognition are two important and challenging problems in many real-world applications and they have drawn increasing attention from the deep learning community in recent year…
Activity RecognitionBenchmarkingHuman Activity RecognitionSegmentationFeature Selection on a Flare Forecasting Testbed: A Comparative Study of 24 Methods
The Space-Weather ANalytics for Solar Flares (SWAN-SF) is a multivariate time series benchmark dataset recently created to serve the heliophysics community as a testbed for solar flare forecasting models. SWAN-SF contain…
feature selectionTime SeriesTime Series Analysis