Subject-dependent and -independent human activity recognition with person-specific and -independent models

Research output: Chapter in Book/Report/Conference proceedingsChapterpeer-review

Abstract

The distinction between subject-dependent and subject-independent performance is ubiquitous in the Human Activity Recognition (HAR) literature. We test the hypotheses that HAR models achieve better subject-dependent performance than subject-independent performance, that a model trained with many users will achieve better subject-independent performance than one trained with a single user, and that one trained with a single user performs better for that user than one trained with this and other users by comparing four algorithms' subject-dependent and -independent performance across eight data sets using three different approaches, which we term person-independent models (PIMs), person-specific models (PSMs), and ensembles of PSMs (EPSMs). Our analysis shows that PSMs outperform PIMs by 3.5% for known users, PIMs outperform PSMs by 13.9% and ensembles of PSMs by a not significant 2.1% for unknown users, and that the performance for known users is 20.5% to 48% better than for unknown users.

Original languageEnglish
Title of host publicationiWOAR 2019 - 6th International Workshop on Sensor-Based Activity Recognition and Interaction, Proceedings
EditorsStefan Ludtke, Sebastian Bader, Kristina Yordanova, Thomas Kirste
PublisherAssociation for Computing Machinery
ISBN (Electronic)9781450377140
DOIs
Publication statusPublished - 16 Sep 2019
Event6th International Workshop on Sensor-Based Activity Recognition and Interaction, iWOAR 2019 - Rostock, Germany
Duration: 16 Sep 201917 Sep 2019

Publication series

NameACM International Conference Proceeding Series

Conference

Conference6th International Workshop on Sensor-Based Activity Recognition and Interaction, iWOAR 2019
Country/TerritoryGermany
CityRostock
Period16/09/1917/09/19

Keywords

  • Bagging
  • Boosting
  • Ensemble Methods
  • Human Activity Recognition
  • Inertial Sensors
  • Machine Learning

Fingerprint

Dive into the research topics of 'Subject-dependent and -independent human activity recognition with person-specific and -independent models'. Together they form a unique fingerprint.

Cite this