Graph-Structure-Based Multigranular Belief Fusion for Human Activity Recognition

The belief functions (BFs) introduced by Shafer in the mid of 1970s are widely applied in information fusion to model epistemic uncertainty and to reason about uncertainty. Their success in applications is however limited because of their high-computational complexity in the fusion process, especial...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transaction on neural networks and learning systems 2024-10, Vol.35 (10), p.13589-13603
Hauptverfasser: Dong, Yilin, Li, Xinde, Dezert, Jean, Zhou, Rigui, Zuo, Kezhu, Ge, Shuzhi Sam
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 13603
container_issue 10
container_start_page 13589
container_title IEEE transaction on neural networks and learning systems
container_volume 35
creator Dong, Yilin
Li, Xinde
Dezert, Jean
Zhou, Rigui
Zuo, Kezhu
Ge, Shuzhi Sam
description The belief functions (BFs) introduced by Shafer in the mid of 1970s are widely applied in information fusion to model epistemic uncertainty and to reason about uncertainty. Their success in applications is however limited because of their high-computational complexity in the fusion process, especially when the number of focal elements is large. To reduce the complexity of reasoning with BFs, we can envisage as a first method to reduce the number of focal elements involved in the fusion process to convert the original basic belief assignments (BBAs) into simpler ones, or as a second method to use a simple rule of combination with potentially a loss of the specificity and pertinence of the fusion result, or to apply both methods jointly. In this article, we focus on the first method and propose a new BBA granulation method inspired by the community clustering of nodes in graph networks. This article studies a novel efficient multigranular belief fusion (MGBF) method. Specifically, focal elements are regarded as nodes in the graph structure, and the distance between nodes will be used to discover the local community relationship of focal elements. Afterward, the nodes belonging to the decision-making community are specially selected, and then the derived multigranular sources of evidence can be efficiently combined. To evaluate the effectiveness of the proposed graph-based MGBF, we further apply this new approach to combine the outputs of convolutional neural networks + attention (CNN + Attention) in the human activity recognition (HAR) problem. The experimental results obtained with real datasets prove the potential interest and feasibility of our proposed strategy with respect to classical BF fusion methods.
doi_str_mv 10.1109/TNNLS.2023.3270290
format Article
fullrecord <record><control><sourceid>proquest_RIE</sourceid><recordid>TN_cdi_ieee_primary_10132427</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>10132427</ieee_id><sourcerecordid>2819278307</sourcerecordid><originalsourceid>FETCH-LOGICAL-c309t-e9c33e600b24af68fcabc5c587c3405be573f397e3477056e74528d676cc89de3</originalsourceid><addsrcrecordid>eNpNkF1LwzAUhoMoTtQ_ICK91IvO5KRt0ss5nBPmB26CdyHLTmeka2fSDPbv7dwc5ibhnOd9CQ8hF4x2GaP57eT5eTTuAgXe5SAo5PSAnADLIAYu5eH-LT465Nz7L9qejKZZkh-TDhcACU_hhLw-OL38jMeNC6YJDuM77XEWPYWysXOnq1BqF91habGIBsHbuoqK2kXDsNBV1DONXdlmHb2hqeeVbdr1GTkqdOnxfHefkvfB_aQ_jEcvD4_93ig2nOZNjLnhHDNKp5DoIpOF0VOTmlQKwxOaTjEVvOC5QJ4I0X4bRZKCnGUiM0bmM-Sn5Gbb-6lLtXR2od1a1dqqYW-kNjOaZJBKyVasZa-37NLV3wF9oxbWGyxLXWEdvALJchCSU9GisEWNq713WOy7GVUb8epXvNqIVzvxbehq1x-mC5ztI3-aW-ByC1hE_NfIOCQg-A_YF4ZA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2819278307</pqid></control><display><type>article</type><title>Graph-Structure-Based Multigranular Belief Fusion for Human Activity Recognition</title><source>IEEE Electronic Library (IEL)</source><creator>Dong, Yilin ; Li, Xinde ; Dezert, Jean ; Zhou, Rigui ; Zuo, Kezhu ; Ge, Shuzhi Sam</creator><creatorcontrib>Dong, Yilin ; Li, Xinde ; Dezert, Jean ; Zhou, Rigui ; Zuo, Kezhu ; Ge, Shuzhi Sam</creatorcontrib><description>The belief functions (BFs) introduced by Shafer in the mid of 1970s are widely applied in information fusion to model epistemic uncertainty and to reason about uncertainty. Their success in applications is however limited because of their high-computational complexity in the fusion process, especially when the number of focal elements is large. To reduce the complexity of reasoning with BFs, we can envisage as a first method to reduce the number of focal elements involved in the fusion process to convert the original basic belief assignments (BBAs) into simpler ones, or as a second method to use a simple rule of combination with potentially a loss of the specificity and pertinence of the fusion result, or to apply both methods jointly. In this article, we focus on the first method and propose a new BBA granulation method inspired by the community clustering of nodes in graph networks. This article studies a novel efficient multigranular belief fusion (MGBF) method. Specifically, focal elements are regarded as nodes in the graph structure, and the distance between nodes will be used to discover the local community relationship of focal elements. Afterward, the nodes belonging to the decision-making community are specially selected, and then the derived multigranular sources of evidence can be efficiently combined. To evaluate the effectiveness of the proposed graph-based MGBF, we further apply this new approach to combine the outputs of convolutional neural networks + attention (CNN + Attention) in the human activity recognition (HAR) problem. The experimental results obtained with real datasets prove the potential interest and feasibility of our proposed strategy with respect to classical BF fusion methods.</description><identifier>ISSN: 2162-237X</identifier><identifier>ISSN: 2162-2388</identifier><identifier>EISSN: 2162-2388</identifier><identifier>DOI: 10.1109/TNNLS.2023.3270290</identifier><identifier>PMID: 37224352</identifier><identifier>CODEN: ITNNAL</identifier><language>eng</language><publisher>United States: IEEE</publisher><subject>Algorithms ; Approximation methods ; Basic belief assignment (BBA) ; belief functions (BFs) ; Computational complexity ; Decision making ; Engineering Sciences ; Evidence theory ; graph networks ; Human Activities ; human activity recognition (HAR) ; Humans ; Learning systems ; Measurement ; multigranular fusion ; Neural Networks, Computer ; Pattern Recognition, Automated - methods ; Physics ; Visualization</subject><ispartof>IEEE transaction on neural networks and learning systems, 2024-10, Vol.35 (10), p.13589-13603</ispartof><rights>Distributed under a Creative Commons Attribution 4.0 International License</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c309t-e9c33e600b24af68fcabc5c587c3405be573f397e3477056e74528d676cc89de3</cites><orcidid>0000-0001-5549-312X ; 0000-0002-8894-8108 ; 0000-0002-4441-3355 ; 0000-0002-1529-4537 ; 0000-0003-3474-9186 ; 0000-0002-9426-1599</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/10132427$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>230,315,781,785,797,886,27929,27930,54763</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/10132427$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/37224352$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink><backlink>$$Uhttps://hal.science/hal-04625881$$DView record in HAL$$Hfree_for_read</backlink></links><search><creatorcontrib>Dong, Yilin</creatorcontrib><creatorcontrib>Li, Xinde</creatorcontrib><creatorcontrib>Dezert, Jean</creatorcontrib><creatorcontrib>Zhou, Rigui</creatorcontrib><creatorcontrib>Zuo, Kezhu</creatorcontrib><creatorcontrib>Ge, Shuzhi Sam</creatorcontrib><title>Graph-Structure-Based Multigranular Belief Fusion for Human Activity Recognition</title><title>IEEE transaction on neural networks and learning systems</title><addtitle>TNNLS</addtitle><addtitle>IEEE Trans Neural Netw Learn Syst</addtitle><description>The belief functions (BFs) introduced by Shafer in the mid of 1970s are widely applied in information fusion to model epistemic uncertainty and to reason about uncertainty. Their success in applications is however limited because of their high-computational complexity in the fusion process, especially when the number of focal elements is large. To reduce the complexity of reasoning with BFs, we can envisage as a first method to reduce the number of focal elements involved in the fusion process to convert the original basic belief assignments (BBAs) into simpler ones, or as a second method to use a simple rule of combination with potentially a loss of the specificity and pertinence of the fusion result, or to apply both methods jointly. In this article, we focus on the first method and propose a new BBA granulation method inspired by the community clustering of nodes in graph networks. This article studies a novel efficient multigranular belief fusion (MGBF) method. Specifically, focal elements are regarded as nodes in the graph structure, and the distance between nodes will be used to discover the local community relationship of focal elements. Afterward, the nodes belonging to the decision-making community are specially selected, and then the derived multigranular sources of evidence can be efficiently combined. To evaluate the effectiveness of the proposed graph-based MGBF, we further apply this new approach to combine the outputs of convolutional neural networks + attention (CNN + Attention) in the human activity recognition (HAR) problem. The experimental results obtained with real datasets prove the potential interest and feasibility of our proposed strategy with respect to classical BF fusion methods.</description><subject>Algorithms</subject><subject>Approximation methods</subject><subject>Basic belief assignment (BBA)</subject><subject>belief functions (BFs)</subject><subject>Computational complexity</subject><subject>Decision making</subject><subject>Engineering Sciences</subject><subject>Evidence theory</subject><subject>graph networks</subject><subject>Human Activities</subject><subject>human activity recognition (HAR)</subject><subject>Humans</subject><subject>Learning systems</subject><subject>Measurement</subject><subject>multigranular fusion</subject><subject>Neural Networks, Computer</subject><subject>Pattern Recognition, Automated - methods</subject><subject>Physics</subject><subject>Visualization</subject><issn>2162-237X</issn><issn>2162-2388</issn><issn>2162-2388</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>RIE</sourceid><sourceid>EIF</sourceid><recordid>eNpNkF1LwzAUhoMoTtQ_ICK91IvO5KRt0ss5nBPmB26CdyHLTmeka2fSDPbv7dwc5ibhnOd9CQ8hF4x2GaP57eT5eTTuAgXe5SAo5PSAnADLIAYu5eH-LT465Nz7L9qejKZZkh-TDhcACU_hhLw-OL38jMeNC6YJDuM77XEWPYWysXOnq1BqF91habGIBsHbuoqK2kXDsNBV1DONXdlmHb2hqeeVbdr1GTkqdOnxfHefkvfB_aQ_jEcvD4_93ig2nOZNjLnhHDNKp5DoIpOF0VOTmlQKwxOaTjEVvOC5QJ4I0X4bRZKCnGUiM0bmM-Sn5Gbb-6lLtXR2od1a1dqqYW-kNjOaZJBKyVasZa-37NLV3wF9oxbWGyxLXWEdvALJchCSU9GisEWNq713WOy7GVUb8epXvNqIVzvxbehq1x-mC5ztI3-aW-ByC1hE_NfIOCQg-A_YF4ZA</recordid><startdate>20241001</startdate><enddate>20241001</enddate><creator>Dong, Yilin</creator><creator>Li, Xinde</creator><creator>Dezert, Jean</creator><creator>Zhou, Rigui</creator><creator>Zuo, Kezhu</creator><creator>Ge, Shuzhi Sam</creator><general>IEEE</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>1XC</scope><scope>VOOES</scope><orcidid>https://orcid.org/0000-0001-5549-312X</orcidid><orcidid>https://orcid.org/0000-0002-8894-8108</orcidid><orcidid>https://orcid.org/0000-0002-4441-3355</orcidid><orcidid>https://orcid.org/0000-0002-1529-4537</orcidid><orcidid>https://orcid.org/0000-0003-3474-9186</orcidid><orcidid>https://orcid.org/0000-0002-9426-1599</orcidid></search><sort><creationdate>20241001</creationdate><title>Graph-Structure-Based Multigranular Belief Fusion for Human Activity Recognition</title><author>Dong, Yilin ; Li, Xinde ; Dezert, Jean ; Zhou, Rigui ; Zuo, Kezhu ; Ge, Shuzhi Sam</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c309t-e9c33e600b24af68fcabc5c587c3405be573f397e3477056e74528d676cc89de3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Algorithms</topic><topic>Approximation methods</topic><topic>Basic belief assignment (BBA)</topic><topic>belief functions (BFs)</topic><topic>Computational complexity</topic><topic>Decision making</topic><topic>Engineering Sciences</topic><topic>Evidence theory</topic><topic>graph networks</topic><topic>Human Activities</topic><topic>human activity recognition (HAR)</topic><topic>Humans</topic><topic>Learning systems</topic><topic>Measurement</topic><topic>multigranular fusion</topic><topic>Neural Networks, Computer</topic><topic>Pattern Recognition, Automated - methods</topic><topic>Physics</topic><topic>Visualization</topic><toplevel>online_resources</toplevel><creatorcontrib>Dong, Yilin</creatorcontrib><creatorcontrib>Li, Xinde</creatorcontrib><creatorcontrib>Dezert, Jean</creatorcontrib><creatorcontrib>Zhou, Rigui</creatorcontrib><creatorcontrib>Zuo, Kezhu</creatorcontrib><creatorcontrib>Ge, Shuzhi Sam</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Electronic Library (IEL)</collection><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>Hyper Article en Ligne (HAL)</collection><collection>Hyper Article en Ligne (HAL) (Open Access)</collection><jtitle>IEEE transaction on neural networks and learning systems</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Dong, Yilin</au><au>Li, Xinde</au><au>Dezert, Jean</au><au>Zhou, Rigui</au><au>Zuo, Kezhu</au><au>Ge, Shuzhi Sam</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Graph-Structure-Based Multigranular Belief Fusion for Human Activity Recognition</atitle><jtitle>IEEE transaction on neural networks and learning systems</jtitle><stitle>TNNLS</stitle><addtitle>IEEE Trans Neural Netw Learn Syst</addtitle><date>2024-10-01</date><risdate>2024</risdate><volume>35</volume><issue>10</issue><spage>13589</spage><epage>13603</epage><pages>13589-13603</pages><issn>2162-237X</issn><issn>2162-2388</issn><eissn>2162-2388</eissn><coden>ITNNAL</coden><abstract>The belief functions (BFs) introduced by Shafer in the mid of 1970s are widely applied in information fusion to model epistemic uncertainty and to reason about uncertainty. Their success in applications is however limited because of their high-computational complexity in the fusion process, especially when the number of focal elements is large. To reduce the complexity of reasoning with BFs, we can envisage as a first method to reduce the number of focal elements involved in the fusion process to convert the original basic belief assignments (BBAs) into simpler ones, or as a second method to use a simple rule of combination with potentially a loss of the specificity and pertinence of the fusion result, or to apply both methods jointly. In this article, we focus on the first method and propose a new BBA granulation method inspired by the community clustering of nodes in graph networks. This article studies a novel efficient multigranular belief fusion (MGBF) method. Specifically, focal elements are regarded as nodes in the graph structure, and the distance between nodes will be used to discover the local community relationship of focal elements. Afterward, the nodes belonging to the decision-making community are specially selected, and then the derived multigranular sources of evidence can be efficiently combined. To evaluate the effectiveness of the proposed graph-based MGBF, we further apply this new approach to combine the outputs of convolutional neural networks + attention (CNN + Attention) in the human activity recognition (HAR) problem. The experimental results obtained with real datasets prove the potential interest and feasibility of our proposed strategy with respect to classical BF fusion methods.</abstract><cop>United States</cop><pub>IEEE</pub><pmid>37224352</pmid><doi>10.1109/TNNLS.2023.3270290</doi><tpages>15</tpages><orcidid>https://orcid.org/0000-0001-5549-312X</orcidid><orcidid>https://orcid.org/0000-0002-8894-8108</orcidid><orcidid>https://orcid.org/0000-0002-4441-3355</orcidid><orcidid>https://orcid.org/0000-0002-1529-4537</orcidid><orcidid>https://orcid.org/0000-0003-3474-9186</orcidid><orcidid>https://orcid.org/0000-0002-9426-1599</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier ISSN: 2162-237X
ispartof IEEE transaction on neural networks and learning systems, 2024-10, Vol.35 (10), p.13589-13603
issn 2162-237X
2162-2388
2162-2388
language eng
recordid cdi_ieee_primary_10132427
source IEEE Electronic Library (IEL)
subjects Algorithms
Approximation methods
Basic belief assignment (BBA)
belief functions (BFs)
Computational complexity
Decision making
Engineering Sciences
Evidence theory
graph networks
Human Activities
human activity recognition (HAR)
Humans
Learning systems
Measurement
multigranular fusion
Neural Networks, Computer
Pattern Recognition, Automated - methods
Physics
Visualization
title Graph-Structure-Based Multigranular Belief Fusion for Human Activity Recognition
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-15T15%3A36%3A09IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_RIE&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Graph-Structure-Based%20Multigranular%20Belief%20Fusion%20for%20Human%20Activity%20Recognition&rft.jtitle=IEEE%20transaction%20on%20neural%20networks%20and%20learning%20systems&rft.au=Dong,%20Yilin&rft.date=2024-10-01&rft.volume=35&rft.issue=10&rft.spage=13589&rft.epage=13603&rft.pages=13589-13603&rft.issn=2162-237X&rft.eissn=2162-2388&rft.coden=ITNNAL&rft_id=info:doi/10.1109/TNNLS.2023.3270290&rft_dat=%3Cproquest_RIE%3E2819278307%3C/proquest_RIE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2819278307&rft_id=info:pmid/37224352&rft_ieee_id=10132427&rfr_iscdi=true