Semi-Automated Data Labeling for Activity Recognition in Pervasive Healthcare
Authors:
- Dagoberto Cruz-SandovalDCDagoberto Cruz-Sandoval
- Jessica Beltran-MarquezJBJessica Beltran-Marquez
- Matias Garcia-ConstantinoMatias Garcia-Constantino
- Luis A. Gonzalez-JassoLGLuis A. Gonzalez-Jasso
- Jesus FavelaJFJesus Favela
- Irvin Hussein Lopez-NavaILIrvin Hussein Lopez-Nava
- Ian ClelandIan Cleland
- Andrew EnnisAEAndrew Ennis
- Netzahualcoyotl Hernandez-CruzNetzahualcoyotl Hernandez-Cruz
- Joseph RaffertyJoseph Rafferty
- Jonathan SynnottJSJonathan Synnott
- Chris NugentChris Nugent
Open Access
Abstract
Activity recognition, a key component in pervasive healthcare monitoring, relies on classification algorithms that require labeled data of individuals performing the activity of interest to train accurate models. Labeling data can be performed in a lab setting where an individual enacts the activity under controlled conditions. The ubiquity of mobile and wearable sensors allows the collection of large datasets from individuals performing activities in naturalistic conditions. Gathering accurate data labels for activity recognition is typically an expensive and time-consuming process. In this paper we present two novel approaches for semi-automated online data labeling performed by the individual executing the activity of interest. The approaches have been designed to address two of the limitations of self-annotation: (i) The burden on the user performing and annotating the activity, and (ii) the lack of accuracy due to the user labeling the data minutes or hours after the completion of an activity. The first approach is based on the recognition of subtle finger gestures performed in response to a data-labeling query. The second approach focuses on labeling activities that have an auditory manifestation and uses a classifier to have an initial estimation of the activity, and a conversational agent to ask the participant for clarification or for additional data. Both approaches are described, evaluated in controlled experiments to assess their feasibility and their advantages and limitations are discussed. Results show that while both studies have limitations, they achieve 80% to 90% precision.Keywords
Funding Information
- UK Research and Innovation (2018-1)
- Secretaría de Ciencia, Humanidades, Tecnología e Innovación (Scholarship No. 276379)
This publication has 30 references indexed in Scilit:
- Scalable identification of mixed environmental sounds, recorded from heterogeneous sourcesPattern Recognition Letters, 2015
- Detection of Gestures Associated With Medication Adherence Using Smartwatch-Based Inertial SensorsIEEE Sensors Journal, 2015
- Automatic annotation of tennis games: An integration of audio, vision, and learningImage and Vision Computing, 2014
- CARMA: Software for Continuous Affect Rating and Media AnnotationJournal of Open Research Software, 2014
- Do Flawed Data on Caloric Intake From NHANES Present Problems for Researchers and Policy Makers?JAMA, 2013
- Multimodal recognition of reading activity in transit using body-worn sensorsACM Transactions on Applied Perception, 2012
- What Is Ecological Validity? A Dimensional AnalysisInfancy, 2001
- MARSYAS: a framework for audio analysisOrganised Sound, 2000
- Automatic segmentation and labelling of multi-lingual speech dataSpeech Communication, 1996
- Self‐Recording of Everyday Life Events: Origins, Types, and UsesJournal of Personality, 1991