Probabilistic Learning from Incomplete Data for Recognition of Activities of Daily Living in Smart Homes

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

Learning behavioral patterns for activities of daily living in a smart home environment can be challenged by the limited number of training data that may be available. This may be due to the infrequent repetition of routine activities (e.g., once daily), the expense of using observers to label activities, and the intrusion that would be caused by the presence of observers over long time periods. It is important, therefore, to make as much use of any labeled data that are collected, however, incomplete these data may be. In this paper, we propose an algorithm for learning behavioral patterns for multi-inhabitants living in a single smart home environment, by making full use of all limited labeled activities, including incomplete data resulting from unreliable low-level sensors in this environment. Through maximum-likelihood estimation, using Expectation-Maximization, we build a model that captures both environmental uncertainties from sensor readings and user uncertainties, including variations in how individuals carry out activities. Our algorithm outperforms models that cannot handle data incompleteness, with increasing performance gains as incompleteness increases. The approach also enables the impact of particular sensors to be assessed and can thus inform sensor maintenance and deployment.
LanguageEnglish
Pages454-462
JournalIEEE Transactions on Information Technology in BioMedicine
Volume16
Issue number3
DOIs
Publication statusPublished - May 2012

Fingerprint

Sensors
Maximum likelihood estimation
Labels
Uncertainty

Keywords

  • Activity recognition
  • activities of daily living (ADLs)
  • Expectation–Maximization (EM) algorithm
  • incomplete data
  • probabilistic learning

Cite this

@article{6ef93875281e4c4796c8f1c0dce24495,
title = "Probabilistic Learning from Incomplete Data for Recognition of Activities of Daily Living in Smart Homes",
abstract = "Learning behavioral patterns for activities of daily living in a smart home environment can be challenged by the limited number of training data that may be available. This may be due to the infrequent repetition of routine activities (e.g., once daily), the expense of using observers to label activities, and the intrusion that would be caused by the presence of observers over long time periods. It is important, therefore, to make as much use of any labeled data that are collected, however, incomplete these data may be. In this paper, we propose an algorithm for learning behavioral patterns for multi-inhabitants living in a single smart home environment, by making full use of all limited labeled activities, including incomplete data resulting from unreliable low-level sensors in this environment. Through maximum-likelihood estimation, using Expectation-Maximization, we build a model that captures both environmental uncertainties from sensor readings and user uncertainties, including variations in how individuals carry out activities. Our algorithm outperforms models that cannot handle data incompleteness, with increasing performance gains as incompleteness increases. The approach also enables the impact of particular sensors to be assessed and can thus inform sensor maintenance and deployment.",
keywords = "Activity recognition, activities of daily living (ADLs), Expectation–Maximization (EM) algorithm, incomplete data, probabilistic learning",
author = "Shuai Zhang and SI McClean and BW Scotney",
year = "2012",
month = "5",
doi = "10.1109/TITB.2012.2188534",
language = "English",
volume = "16",
pages = "454--462",
journal = "IEEE Transactions on Information Technology in BioMedicine",
issn = "1089-7771",
number = "3",

}

TY - JOUR

T1 - Probabilistic Learning from Incomplete Data for Recognition of Activities of Daily Living in Smart Homes

AU - Zhang, Shuai

AU - McClean, SI

AU - Scotney, BW

PY - 2012/5

Y1 - 2012/5

N2 - Learning behavioral patterns for activities of daily living in a smart home environment can be challenged by the limited number of training data that may be available. This may be due to the infrequent repetition of routine activities (e.g., once daily), the expense of using observers to label activities, and the intrusion that would be caused by the presence of observers over long time periods. It is important, therefore, to make as much use of any labeled data that are collected, however, incomplete these data may be. In this paper, we propose an algorithm for learning behavioral patterns for multi-inhabitants living in a single smart home environment, by making full use of all limited labeled activities, including incomplete data resulting from unreliable low-level sensors in this environment. Through maximum-likelihood estimation, using Expectation-Maximization, we build a model that captures both environmental uncertainties from sensor readings and user uncertainties, including variations in how individuals carry out activities. Our algorithm outperforms models that cannot handle data incompleteness, with increasing performance gains as incompleteness increases. The approach also enables the impact of particular sensors to be assessed and can thus inform sensor maintenance and deployment.

AB - Learning behavioral patterns for activities of daily living in a smart home environment can be challenged by the limited number of training data that may be available. This may be due to the infrequent repetition of routine activities (e.g., once daily), the expense of using observers to label activities, and the intrusion that would be caused by the presence of observers over long time periods. It is important, therefore, to make as much use of any labeled data that are collected, however, incomplete these data may be. In this paper, we propose an algorithm for learning behavioral patterns for multi-inhabitants living in a single smart home environment, by making full use of all limited labeled activities, including incomplete data resulting from unreliable low-level sensors in this environment. Through maximum-likelihood estimation, using Expectation-Maximization, we build a model that captures both environmental uncertainties from sensor readings and user uncertainties, including variations in how individuals carry out activities. Our algorithm outperforms models that cannot handle data incompleteness, with increasing performance gains as incompleteness increases. The approach also enables the impact of particular sensors to be assessed and can thus inform sensor maintenance and deployment.

KW - Activity recognition

KW - activities of daily living (ADLs)

KW - Expectation–Maximization (EM) algorithm

KW - incomplete data

KW - probabilistic learning

U2 - 10.1109/TITB.2012.2188534

DO - 10.1109/TITB.2012.2188534

M3 - Article

VL - 16

SP - 454

EP - 462

JO - IEEE Transactions on Information Technology in BioMedicine

T2 - IEEE Transactions on Information Technology in BioMedicine

JF - IEEE Transactions on Information Technology in BioMedicine

SN - 1089-7771

IS - 3

ER -