Ontology-based feature generation to improve accuracy of activity recognition in smart environments
Computers & Electrical EngineeringPublished 3 April 2018Open access
Alberto Salguero, Macarena Espinilla
Citations21
SJR quartileQ1
SJR score1.05
SNIP1.63
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
The use of ontologies is proposed for the fully automatic generation of features of Activities of Daily Living from smart home sensor data by converting the original dataset into an ontology and then combining all the concepts and relations in that ontology to obtain relevant class expressions.
Abstract
<p>In recent years, many techniques have been proposed for automatic recognition of Activities ofDaily Living from smart home sensor data. However, classifiers usually use features created adhoc. In this work, the use of ontologies is proposed for the fully automatic generation of thesefeatures. The process consists of converting the original dataset into an ontology and thencombine all the concepts and relations in that ontology to obtain relevant class expressions. Thehigh formalization of ontologies allows us to reduce the search space by discarding manymeaningless expressions, such as contradictory or unsatisfiable expressions. The relevant classexpressions are then used as features by the classifiers to build the classification model. To va-lidate our proposal, we have used as reference the results obtained by four different classificationalgorithms that use the most commonly used features.</p>
Keywords
Computer Science
Accurate activity recognition in a home setting
869 Citations2008Tim van Kasteren, A. Noulas +2 more
This paper presents an easy to install sensor network and an accurate but inexpensive annotation method and shows how the hidden Markov model and conditional random fields perform in recognizing activities.
Proceedings of the 19th ACM international conference on Information and knowledge management
620 Citations2010
This year CIKM has received a record high number of submissions, as can be seen from the following statistics: 1382 abstracts submitted, 945 full papers plus 38 demo papers submitted, and 126 papers accepted for presentation as full papers and an additional 165 accepted for short papers.
IEEE Transactions on Knowledge and Data EngineeringA Knowledge-Driven Approach to Activity Recognition in Smart Homes
582 Citations2011Liming Chen, Chris Nugent +1 more
This paper presents a generic system architecture for the proposed knowledge-driven approach to real-time, continuous activity recognition based on multisensor data streams in smart homes, and describes the underlying ontology-based recognition process.
IEEE Transactions on Pattern Analysis and Machine IntelligenceDiscriminative Learning of Local Image Descriptors
478 Citations2010Matthew A. Brown, Gang Hua +1 more
A set of building blocks for constructing descriptors which can be combined together and jointly optimized so as to minimize the error of a nearest-neighbor classifier are described.
SensorsActivity Recognition Using Hybrid Generative/Discriminative Models on Home Environments Using Binary Sensors
278 Citations2013Fco. Javier Ordóñez, Paula de Toledo +1 more
This paper describes the use of two powerful machine learning schemes, ANN and SVM, within the framework of HMM (Hidden Markov Model), in order to tackle the task of activity recognition in a home setting and shows how the hybrid models achieve significantly better recognition performance.
IEEE Transactions on Human-Machine SystemsAn Ontology-Based Hybrid Approach to Activity Modeling for Smart Homes
143 Citations2014Liming Chen, Chris Nugent +1 more
An ontology-based hybrid approach to activity modeling that combines domain knowledge based model specification and data-driven model learning is introduced that has been implemented in a feature-rich assistive living system.
Journal of Web SemanticsClass expression learning for ontology engineering
134 Citations2011Jens Lehmann, Sören Auer +2 more
This article describes how to adapt a semi-automatic method for learning OWL class expressions to the ontology engineering use case and performs rigorous performance optimization of the underlying algorithms for providing instant suggestions to the user.
Putting OWL in order: Patterns for sequences in OWL
68 Citations2006Nick Drummond, Alan Rector +5 more
This paper demonstrates a design pattern for modeling order within OWL-DL that allows us to use standard DL reasoning to perform pattern matching akin to regular expression matching and works surprisingly well.
PubMedTracking Activities in Complex Settings Using Smart Environment Technologies.
67 Citations2009Geetika Singla, Diane J. Cook +1 more
This paper looks at approaches to perform real-time recognition of Activities of Daily Living, and enhances other related research efforts to develop approaches that are effective when activities are interrupted and interleaved.
Lecture notes in computer scienceGenerating Possible Interpretations for Statistics from Linked Open Data
61 Citations2012Heiko Paulheim
This paper introduces Explain-a-LOD, an approach which uses data from Linked Open Data for generating hypotheses that explain statistics, and shows an implemented prototype and compares different approaches for generating hypothesis by analyzing the perceived quality of those hypotheses in a user study.
Automated feature generation from structured knowledge
52 Citations2011Weiwei Cheng, Gjergji Kasneci +3 more
This paper introduces an expressive graph-based language for extracting features from such knowledge bases and a theoretical framework for constructing feature vectors from the extracted features.
Future Generation Computer SystemsAn ontology-based approach to ADL recognition in smart homes
52 Citations2013Ihn-Han Bae
The ontology base and the reasoning that are the main parts of ADL pattern monitoring are studied, which can provide unprecedented opportunities for health monitoring and assisted living applications, especially for elderly people and people with memory deficits.
Knowledge-Based SystemsEnhancing ontological reasoning with uncertainty handling for activity recognition
42 Citations2016Mohd Halim Mohd Noor, Zoran Salčić +1 more
Results have shown that the proposed reasoning approach can accommodate uncertainties and subsequently infer the activities more accurately in comparison with existing ontology-based recognition and perform comparably well to the data-driven approach.
Microprocessors and MicrosystemsOptimizing the configuration of an heterogeneous architecture of sensors for activity recognition, using the extended belief rule-based inference methodology
36 Citations2016Macarena Espinilla, Javier Medina-Quero +4 more
This paper proposes the adaption of the extended belief rule-based inference methodology (RIMER+) to handle data binary sensors and its use as the suitable classifier for activity recognition that keeps the accuracy of results even in situations where an essential sensor fails.
Knowledge-Based SystemsUncertainty handling in semantic reasoning for accurate context understanding
33 Citations2015Hamdi Aloulou, Mounir Mokhtari +3 more
This paper proposes a method for the measurement of uncertainty based on both physical and operational behaviors of the sensors and describes how the level of uncertainty is incorporated into different layers of a semantically driven context aware system and how it is transferred to a decision engine in order to perform more accurate decisions in ambiguous observations.
The Journal of SupercomputingIntegrating the enriched feature with machine learning algorithms for human movement and fall detection
32 Citations2013Chenghua Li, Man Lin +2 more
Methods to process the human motion data received from wearable wireless sensor devices for detecting different types of human movement activities such as sitting, standing, lying, fall, running, and walking are studied.
Journal of Ambient Intelligence and Humanized ComputingInterpreting human activity from electrical consumption data using reconfigurable hardware and hidden Markov models
29 Citations2016Francisco Javier Ferrández Pastor, Higinio Mora +3 more
Non-intrusive load monitoring, design based on Wavelet Transform processing and Field Programmable Gate Arrays hardware implementation provide the necessary support to develop this kind of embedded systems.
Goal Lifecycles and Ontological Models for Intention Based Assistive Living within Smart Environments
16 Citations2015Joseph Rafferty, Liming Chen +2 more
A novel approach to assistive living is introduced which intends to address issues with scalability and ability to model activity variations via a paradigm shift from a sensor centric approach to a goal-oriented one.
Computers & Electrical EngineeringHuman activity recognition based on multiple order temporal information
13 Citations2014Jianqin Yin, Guohui Tian +2 more
This paper model the problem of human activity recognition as a classification problem through the definition of a representation scheme that uses multiple order temporal information and shows that the features with low and high support have limited discriminative power.
An initiative for the creation of open datasets within pervasive healthcare
13 Citations2016Chris Nugent, Ian Cleland +7 more
Recommendations of how this can be made a reality are presented in addition to the initial steps which have been taken to facilitate such an initiative involving the definition of common formats for data storage and a common set of tools for data processing and visualization.
Lecture notes in computer scienceRule-Based Real-Time ADL Recognition in a Smart Home Environment
12 Citations2016George Baryannis, Przemyslaw Woznowski +1 more
Novel aspects of the approach include the ability to recognise arbitrary scenarios of complex activities using bottom-up multi-level reasoning, starting from sensor events at the lowest level; an effective heuristics-based method for distinguishing between actual and ghost images in video data; and a highly accurate indoor localisation approach that fuses different sources of location information.
SensorsTowards A Self Adaptive System for Social Wellness
11 Citations2016Asad Masood Khattak, Wajahat Ali Khan +3 more
An ontology-based approach is proposed, where all the daily life activities and patient profile information are modeled in ontology and an inference mechanism that enables fine-grained situation analysis for personalized service recommendations is exploited.
Journal of Ambient Intelligence and Humanized ComputingExtended Body-Angles Algorithm to recognize activities within intelligent environments
9 Citations2017Carlos Gutiérrez López de la Franca, Ramón Hervás +3 more
The main motivation of the present work is to propose the named Extended Body-Angles Algorithm, a method able to recognize movements through single samples of modelled movements using Kinect, based on the Body- Angles Al algorithm.