LLM4TAP: LLM-Enhanced TAP Rule Recommendation

Trigger-action programming (TAP) is an Internet of Things (IoT) paradigm that enables nonprofessional end-users to automate smart devices by formulating rules, such as "IF you leave home, THEN turn off lights." As the number of possible rules increases, manually browsing these rules become...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE internet of things journal Ročník 12; číslo 10; s. 13157 - 13169
Hlavní autoři: Wu, Gang, Hu, Liang, Hu, Yuxiao, Xiong, Xingbo, Wang, Feng
Médium: Journal Article
Jazyk:angličtina
Vydáno: Piscataway IEEE 15.05.2025
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Accuracy
Automation
Cognition
Contrastive learning
Internet of Things
Internet of Things (IoT)
Large language models
large language models (LLMs)
Performance evaluation
Programming
Recommender systems
Representations
rule recommendation
Singular value decomposition
Smart devices
trigger-action programming (TAP)
Turning
ISSN:2327-4662, 2327-4662
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Trigger-action programming (TAP) is an Internet of Things (IoT) paradigm that enables nonprofessional end-users to automate smart devices by formulating rules, such as "IF you leave home, THEN turn off lights." As the number of possible rules increases, manually browsing these rules becomes increasingly time-consuming for users. Recently, graph-based recommendation systems have shown promise in automatically suggesting rules, yet they face two issues. First, these studies struggle to identify and differentiate users' demands (e.g., turning off lights) and intentions (e.g., energy saving). Second, they overlook the issue of sparse user-rule interactions. In this article, we propose LLM4TAP, a large language model (LLM) enhanced TAP rule recommendation framework, to address these issues. Prior to LLM4TAP, a user-rule graph is constructed to represent the interactions between users and rules. Within LLM4TAP, singular value decomposition is first employed to generate an augmented graph, strengthening global structural relationships between users and rules. Next, the reasoning capabilities of LLMs are utilized to infer users' demands and intentions from the textual descriptions of rules and user-rule interactions, producing representations of these inferred demands and intentions. Finally, a dual representation alignment method is introduced, integrating user demands and intentions derived from LLMs with the global structural information from the augmentation graph within a contrastive learning framework to enhance representation performance. Extensive experiments demonstrate the effectiveness of LLM4TAP, achieving the maximum improvements of 8.96% and 4.72% over the strongest compared methods on the IFTTT and Wyze datasets, respectively.
AbstractList Trigger-action programming (TAP) is an Internet of Things (IoT) paradigm that enables nonprofessional end-users to automate smart devices by formulating rules, such as “IF you leave home, THEN turn off lights.” As the number of possible rules increases, manually browsing these rules becomes increasingly time-consuming for users. Recently, graph-based recommendation systems have shown promise in automatically suggesting rules, yet they face two issues. First, these studies struggle to identify and differentiate users’ demands (e.g., turning off lights) and intentions (e.g., energy saving). Second, they overlook the issue of sparse user-rule interactions. In this article, we propose LLM4TAP, a large language model (LLM) enhanced TAP rule recommendation framework, to address these issues. Prior to LLM4TAP, a user-rule graph is constructed to represent the interactions between users and rules. Within LLM4TAP, singular value decomposition is first employed to generate an augmented graph, strengthening global structural relationships between users and rules. Next, the reasoning capabilities of LLMs are utilized to infer users’ demands and intentions from the textual descriptions of rules and user-rule interactions, producing representations of these inferred demands and intentions. Finally, a dual representation alignment method is introduced, integrating user demands and intentions derived from LLMs with the global structural information from the augmentation graph within a contrastive learning framework to enhance representation performance. Extensive experiments demonstrate the effectiveness of LLM4TAP, achieving the maximum improvements of 8.96% and 4.72% over the strongest compared methods on the IFTTT and Wyze datasets, respectively.
Author Hu, Liang
Xiong, Xingbo
Wu, Gang
Hu, Yuxiao
Wang, Feng
Author_xml – sequence: 1
  givenname: Gang
  orcidid: 0000-0001-8721-6995
  surname: Wu
  fullname: Wu, Gang
  email: wugang17@mails.jlu.edu.cn
  organization: College of Computer Science and Technology, Jilin University, Changchun, China
– sequence: 2
  givenname: Liang
  orcidid: 0000-0002-6077-1873
  surname: Hu
  fullname: Hu, Liang
  organization: College of Computer Science and Technology, Jilin University, Changchun, China
– sequence: 3
  givenname: Yuxiao
  surname: Hu
  fullname: Hu, Yuxiao
  organization: College of Computer Science and Technology, Jilin University, Changchun, China
– sequence: 4
  givenname: Xingbo
  surname: Xiong
  fullname: Xiong, Xingbo
  email: xxb_1967@163.com
  organization: College of Computer Science and Technology, Changchun University, Changchun, China
– sequence: 5
  givenname: Feng
  orcidid: 0000-0002-0732-7343
  surname: Wang
  fullname: Wang, Feng
  email: wangfeng12@mails.jlu.edu.cn
  organization: College of Computer Science and Technology, Jilin University, Changchun, China
BookMark eNp9kEtPAjEUhRuDiYj8ABMXk7ge7Iu2444QVMwYDMF108edOAQ62BkW_nuLsCAuXN2Tm_Pdx7lGvdAEQOiW4BEhuHh4nS9WI4rpeMTGjBZSXqA-ZVTmXAjaO9NXaNi2a4xxwsakEH2Ul-UbX03eH7Mk8ln4NMGBz1InW-43kC3BNdstBG-6ugk36LIymxaGpzpAH0-z1fQlLxfP8-mkzB0teJdT4FY65T1YV4DzlHhmrePAsbXceyYlAcOVddhUWLECV4oojy046iXhbIDuj3N3sfnaQ9vpdbOPIa3ULP1JlBAKJ5c8ulxs2jZCpV3d_d7ZRVNvNMH6EI8-xKMP8ehTPIkkf8hdrLcmfv_L3B2ZGgDO_EoKIhj7ATfacEY
CODEN IITJAU
CitedBy_id crossref_primary_10_1109_JIOT_2025_3579920
crossref_primary_10_1007_s10462_025_11320_9
Cites_doi 10.1145/3616855.3635853
10.1109/TMC.2024.3350886
10.1145/3589334.3645682
10.1145/3576842.3582328
10.3390/s24186151
10.1145/3627673.3679564
10.1162/tacl_a_00619
10.1109/JIOT.2021.3103320
10.1145/3589334.3645458
10.1145/3344211
10.1109/TII.2021.3092774
10.1145/3404835.3462862
10.1145/3539618.3591723
10.1145/3534678.3539253
10.1145/3447264
10.1145/3488560.3498527
10.1609/aaai.v33i01.33012547
10.1109/MWC.013.2300485
10.1109/JSAC.2024.3459029
10.1145/3696662
10.26599/TST.2023.9010071
10.1016/j.ijhcs.2018.12.008
10.1016/j.asoc.2024.112163
10.1109/TCOMM.2023.3240697
10.1145/3678585
10.1145/3539618.3591665
10.1007/s00779-024-01825-5
10.1109/tmc.2024.3502685
10.1016/j.eswa.2023.121065
10.1145/3131365.3131369
10.1109/TITS.2024.3418864
10.1145/3637528.3671984
10.1145/3485447.3512104
10.1145/3524610.3527922
10.1145/3604915.3608857
10.1145/3397271.3401063
ContentType Journal Article
Copyright Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025
Copyright_xml – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025
DBID 97E
RIA
RIE
AAYXX
CITATION
7SC
8FD
JQ2
L7M
L~C
L~D
DOI 10.1109/JIOT.2025.3532977
DatabaseName IEEE All-Society Periodicals Package (ASPP) 2005-present
IEEE All-Society Periodicals Package (ASPP) 1998-Present
IEEE Electronic Library Online
CrossRef
Computer and Information Systems Abstracts
Technology Research Database
ProQuest Computer Science Collection
Advanced Technologies Database with Aerospace
Computer and Information Systems Abstracts – Academic
Computer and Information Systems Abstracts Professional
DatabaseTitle CrossRef
Computer and Information Systems Abstracts
Technology Research Database
Computer and Information Systems Abstracts – Academic
Advanced Technologies Database with Aerospace
ProQuest Computer Science Collection
Computer and Information Systems Abstracts Professional
DatabaseTitleList Computer and Information Systems Abstracts
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISSN 2327-4662
EndPage 13169
ExternalDocumentID 10_1109_JIOT_2025_3532977
10876163
Genre orig-research
GrantInformation_xml – fundername: Science and Technology Development Plan of Jilin Province of China
  grantid: 20220101115JC
  funderid: 10.13039/100016694
GroupedDBID 0R~
6IK
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABJNI
ABQJQ
ABVLG
AGQYO
AHBIQ
AKJIK
AKQYR
ALMA_UNASSIGNED_HOLDINGS
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
EBS
IFIPE
IPLJI
JAVBF
M43
OCL
PQQKQ
RIA
RIE
AAYXX
CITATION
7SC
8FD
JQ2
L7M
L~C
L~D
ID FETCH-LOGICAL-c294t-2e4b7c8ddebc9ecd21d3bbc4e40bb4dd3771ea48bc0af08390f818d0bec2d7143
IEDL.DBID RIE
ISICitedReferencesCount 4
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001485409200032&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2327-4662
IngestDate Thu Nov 20 16:51:35 EST 2025
Sat Nov 29 07:58:11 EST 2025
Tue Nov 18 20:51:55 EST 2025
Wed Aug 27 01:53:09 EDT 2025
IsPeerReviewed false
IsScholarly true
Issue 10
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
https://doi.org/10.15223/policy-029
https://doi.org/10.15223/policy-037
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c294t-2e4b7c8ddebc9ecd21d3bbc4e40bb4dd3771ea48bc0af08390f818d0bec2d7143
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ORCID 0000-0002-0732-7343
0000-0001-8721-6995
0000-0002-6077-1873
PQID 3202186680
PQPubID 2040421
PageCount 13
ParticipantIDs crossref_primary_10_1109_JIOT_2025_3532977
ieee_primary_10876163
crossref_citationtrail_10_1109_JIOT_2025_3532977
proquest_journals_3202186680
PublicationCentury 2000
PublicationDate 2025-05-15
PublicationDateYYYYMMDD 2025-05-15
PublicationDate_xml – month: 05
  year: 2025
  text: 2025-05-15
  day: 15
PublicationDecade 2020
PublicationPlace Piscataway
PublicationPlace_xml – name: Piscataway
PublicationTitle IEEE internet of things journal
PublicationTitleAbbrev JIoT
PublicationYear 2025
Publisher IEEE
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml – name: IEEE
– name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References ref13
ref35
ref12
ref34
ref15
ref14
ref36
ref30
ref11
ref33
ref32
ref1
ref16
Kingma (ref39) 2017
ref19
Cai (ref10)
ref18
Rendle (ref37)
Wang (ref17) 2024
Wang (ref2) 2024
ref24
ref23
ref26
ref25
ref20
ref42
ref41
ref22
ref44
ref21
ref43
Asl (ref31)
ref28
ref27
ref29
ref8
ref7
ref9
ref4
ref3
ref6
ref5
Kamani (ref40)
Oord (ref38) 2019
References_xml – ident: ref30
  doi: 10.1145/3616855.3635853
– ident: ref16
  doi: 10.1109/TMC.2024.3350886
– ident: ref28
  doi: 10.1145/3589334.3645682
– ident: ref7
  doi: 10.1145/3576842.3582328
– ident: ref41
  doi: 10.3390/s24186151
– ident: ref33
  doi: 10.1145/3627673.3679564
– ident: ref27
  doi: 10.1162/tacl_a_00619
– ident: ref1
  doi: 10.1109/JIOT.2021.3103320
– ident: ref29
  doi: 10.1145/3589334.3645458
– year: 2019
  ident: ref38
  article-title: Representation learning with contrastive predictive coding
  publication-title: arXiv:1807.03748
– ident: ref22
  doi: 10.1145/3344211
– ident: ref23
  doi: 10.1109/TII.2021.3092774
– ident: ref34
  doi: 10.1145/3404835.3462862
– ident: ref43
  doi: 10.1145/3539618.3591723
– ident: ref42
  doi: 10.1145/3534678.3539253
– ident: ref6
  doi: 10.1145/3447264
– ident: ref35
  doi: 10.1145/3488560.3498527
– ident: ref4
  doi: 10.1609/aaai.v33i01.33012547
– ident: ref18
  doi: 10.1109/MWC.013.2300485
– ident: ref14
  doi: 10.1109/JSAC.2024.3459029
– ident: ref44
  doi: 10.1145/3696662
– year: 2024
  ident: ref2
  article-title: Optimizing 6G integrated sensing and communications (ISAC) via expert networks
  publication-title: arXiv:2406.00408
– ident: ref32
  doi: 10.26599/TST.2023.9010071
– start-page: 452
  volume-title: Proc. 25th Conf. Uncertain. Artif. Intell.
  ident: ref37
  article-title: BPR: Bayesian personalized ranking from implicit feedback
– start-page: 1
  volume-title: Proc. 37th Adv. Neural Inf. Process. Syst.
  ident: ref40
  article-title: Wyze rule: Federated rule Dataset for rule recommendation benchmarking
– ident: ref8
  doi: 10.1016/j.ijhcs.2018.12.008
– year: 2017
  ident: ref39
  article-title: Adam: A method for stochastic optimization
  publication-title: arXiv:1412.6980
– ident: ref11
  doi: 10.1016/j.asoc.2024.112163
– year: 2024
  ident: ref17
  article-title: Generative AI based secure wireless sensing for ISAC networks
  publication-title: arXiv:2408.11398
– ident: ref15
  doi: 10.1109/TCOMM.2023.3240697
– ident: ref20
  doi: 10.1145/3678585
– ident: ref12
  doi: 10.1145/3539618.3591665
– start-page: 1
  volume-title: Proc. 11th Int. Conf. Learn. Represent.
  ident: ref10
  article-title: LightGCL: Simple yet effective graph contrastive learning for recommendation
– ident: ref21
  doi: 10.1007/s00779-024-01825-5
– ident: ref19
  doi: 10.1109/tmc.2024.3502685
– ident: ref24
  doi: 10.1016/j.eswa.2023.121065
– start-page: 3795
  volume-title: Proc. Find. Assoc. Comput. Linguist. NAACL
  ident: ref31
  article-title: RobustSentEmbed: Robust sentence embeddings using adversarial self-supervised contrastive learning
– ident: ref3
  doi: 10.1145/3131365.3131369
– ident: ref13
  doi: 10.1109/TITS.2024.3418864
– ident: ref25
  doi: 10.1145/3637528.3671984
– ident: ref36
  doi: 10.1145/3485447.3512104
– ident: ref5
  doi: 10.1145/3524610.3527922
– ident: ref26
  doi: 10.1145/3604915.3608857
– ident: ref9
  doi: 10.1145/3397271.3401063
SSID ssj0001105196
Score 2.371979
Snippet Trigger-action programming (TAP) is an Internet of Things (IoT) paradigm that enables nonprofessional end-users to automate smart devices by formulating rules,...
SourceID proquest
crossref
ieee
SourceType Aggregation Database
Enrichment Source
Index Database
Publisher
StartPage 13157
SubjectTerms Accuracy
Automation
Cognition
Contrastive learning
Internet of Things
Internet of Things (IoT)
Large language models
large language models (LLMs)
Performance evaluation
Programming
Recommender systems
Representations
rule recommendation
Singular value decomposition
Smart devices
trigger-action programming (TAP)
Turning
Title LLM4TAP: LLM-Enhanced TAP Rule Recommendation
URI https://ieeexplore.ieee.org/document/10876163
https://www.proquest.com/docview/3202186680
Volume 12
WOSCitedRecordID wos001485409200032&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVIEE
  databaseName: IEEE Electronic Library (IEL)
  customDbUrl:
  eissn: 2327-4662
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001105196
  issn: 2327-4662
  databaseCode: RIE
  dateStart: 20140101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEB5s8eDF-qhYrbIHT0JqNps2G29FWlRqLVKht2XzWBTqVvrw9zvJbn0gCt7CkoTlm53MTGZnPoCzsINhh5aUiDClhKNJIZIbQ4SRjCojraUF2YQYDuPJRI7KYnVfC2Ot9T-f2ZYb-ly-memVuypDDUfdRQeiAhUhRFGs9XmhEjpvpFNmLkMqL25v7scYAbJ2K2pHDB2db7bHk6n8OIG9WenX_vlCO7Bd-o9BtxD4LmzYfA9qa26GoFTVfSCDwR0fd0eXAQ5IL3_ymf4AnwQPq6kNXNj5grsXlEp1eOz3xlfXpKRGIJpJviTMciV0jGeT0tJqw0ITKaW55VQpxDoSIrQpj5WmaYZelqQZWmZDUWLMOMrzA6jms9weQtARlGUp78Q0YyirKM64jbIIhasY16lqAF2Dluiyb7ijr5gmPn6gMnE4Jw7npMS5AecfS16Lphl_Ta47YL9MLDBtQHMtmqTUq0Xi2N5di76YHv2y7Bi23O4uwR-2m1Bdzlf2BDb12_J5MT_1n8w79le9LQ
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8NAEB60CnqxPipWq-bgSdi62WybrLciLa2mtUiE3kL2ERRqK334-51NUh-Igrcl7Cbhm8zOzE5mPoALt4lhhxKU-G5CCUeTQgTXmvhaMCq1MIbmZBP-YBCMRmJYFKtntTDGmOznM1O3wyyXr6dqaY_KUMNRd9GBWIeNBufMzcu1Po9UXOuPNIvcpUvF1W3vPsIYkDXqXsNj6Op8sz4ZncqPPTgzLJ3yP19pF3YKD9Jp5SLfgzUz2Yfyip3BKZT1AEgY9nnUGl47OCDtyVOW63fwivOwHBvHBp4vePecVKkCj512dNMlBTkCUUzwBWGGS18FuDtJJYzSzNWelIobTqVEtD3fd03CA6lokqKfJWiKtllTlBnTlvT8EEqT6cQcgdP0KUsT3gxoylBaXpBy46UeilcyrhJZBboCLVZF53BLYDGOswiCitjiHFuc4wLnKlx-LHnN22b8Nbligf0yMce0CrWVaOJCs-ax5Xu3TfoCevzLsnPY6kb9MA57g7sT2LZPsul-t1GD0mK2NKewqd4Wz_PZWfb5vAP6tMB0
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=LLM4TAP%3A+LLM-Enhanced+TAP+Rule+Recommendation&rft.jtitle=IEEE+internet+of+things+journal&rft.au=Wu%2C+Gang&rft.au=Hu%2C+Liang&rft.au=Hu%2C+Yuxiao&rft.au=Xiong%2C+Xingbo&rft.date=2025-05-15&rft.pub=IEEE&rft.eissn=2327-4662&rft.volume=12&rft.issue=10&rft.spage=13157&rft.epage=13169&rft_id=info:doi/10.1109%2FJIOT.2025.3532977&rft.externalDocID=10876163
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2327-4662&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2327-4662&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2327-4662&client=summon