Induction of Non-monotonic Logic Programs To Explain Statistical Learning Models

We present a fast and scalable algorithm to induce non-monotonic logic programs from statistical learning models. We reduce the problem of search for best clauses to instances of the High-Utility Itemset Mining (HUIM) problem. In the HUIM problem, feature values and their importance are treated as t...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Electronic proceedings in theoretical computer science Ročník 306; číslo Proc. ICLP 2019; s. 379 - 388
Hlavný autor: Shakerin, Farhad
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Open Publishing Association 19.09.2019
ISSN:2075-2180, 2075-2180
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract We present a fast and scalable algorithm to induce non-monotonic logic programs from statistical learning models. We reduce the problem of search for best clauses to instances of the High-Utility Itemset Mining (HUIM) problem. In the HUIM problem, feature values and their importance are treated as transactions and utilities respectively. We make use of TreeExplainer, a fast and scalable implementation of the Explainable AI tool SHAP, to extract locally important features and their weights from ensemble tree models. Our experiments with UCI standard benchmarks suggest a significant improvement in terms of classification evaluation metrics and running time of the training algorithm compared to ALEPH, a state-of-the-art Inductive Logic Programming (ILP) system.
AbstractList We present a fast and scalable algorithm to induce non-monotonic logic programs from statistical learning models. We reduce the problem of search for best clauses to instances of the High-Utility Itemset Mining (HUIM) problem. In the HUIM problem, feature values and their importance are treated as transactions and utilities respectively. We make use of TreeExplainer, a fast and scalable implementation of the Explainable AI tool SHAP, to extract locally important features and their weights from ensemble tree models. Our experiments with UCI standard benchmarks suggest a significant improvement in terms of classification evaluation metrics and running time of the training algorithm compared to ALEPH, a state-of-the-art Inductive Logic Programming (ILP) system.
Author Shakerin, Farhad
Author_xml – sequence: 1
  givenname: Farhad
  surname: Shakerin
  fullname: Shakerin, Farhad
  organization: The University of Texas at Dallas
BookMark eNpNkNtKAzEQhoNUsNbe-QD7AG7Nsbu5lFK1sGqh9TrM5lC2bJOSrKBvb2xFnIuZn__iY_iu0cgHbxG6JXjGKeb3y_V2sZkxPJ8JcoHGFFeipKTGo3_5Ck1T2uM8TFIu52O0XnnzoYcu-CK44jX48hB8GILvdNGEXd7rGHYRDqnYhmL5eeyh88VmgKFLQ6ehLxoL0Xd-V7wEY_t0gy4d9MlOf-8EvT8ut4vnsnl7Wi0emlJTUQ-lNJWurTAAtZPcYGmwax2xhBgncxRWSCCMAgegtpKStq7WLXWMQYWhYhO0OnNNgL06xu4A8UsF6NSpCHGnIOYPe6t4ppnWWcmBcZDZAzcOaikx1sAEzqy7M0vHkFK07o9HsPqRq05yVZarBGHfz5pvXQ
Cites_doi 10.1609/aaai.v33i01.33013052
10.1007/978-3-319-57959-7
10.1017/S1471068417000333
10.23919/MIPRO.2018.8400040
10.1145/2939672.2939785
10.14778/2735508.2735510
10.1007/BF03037089
10.1145/2939672.2939778
10.1017/CBO9780511543357
10.2139/ssrn.3063289
10.1007/s10994-011-5259-2
10.1007/BF00117105
10.1016/j.tcs.2005.07.003
10.1145/1055686.1055687
10.1017/CBO9781139342124
ContentType Journal Article
DBID AAYXX
CITATION
DOA
DOI 10.4204/EPTCS.306.51
DatabaseName CrossRef
DOAJ Directory of Open Access Journals
DatabaseTitle CrossRef
DatabaseTitleList
Database_xml – sequence: 1
  dbid: DOA
  name: DOAJ Directory of Open Access Journals
  url: https://www.doaj.org/
  sourceTypes: Open Website
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISSN 2075-2180
EndPage 388
ExternalDocumentID oai_doaj_org_article_4b5edbfe94a34a91804dfa89900ca350
10_4204_EPTCS_306_51
GroupedDBID 5VS
AAYXX
ADBBV
ALMA_UNASSIGNED_HOLDINGS
BCNDV
CITATION
E3Z
GROUPED_DOAJ
KQ8
M~E
OK1
ID FETCH-LOGICAL-c258t-9d7c8e5daa8f94d09d0fbf1e11df90fb5e59a132a4aa2e7992bf8cb2f33a70a73
IEDL.DBID DOA
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001050029200022&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2075-2180
IngestDate Fri Oct 03 12:51:24 EDT 2025
Sat Nov 29 02:05:29 EST 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed false
IsScholarly true
Issue Proc. ICLP 2019
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c258t-9d7c8e5daa8f94d09d0fbf1e11df90fb5e59a132a4aa2e7992bf8cb2f33a70a73
OpenAccessLink https://doaj.org/article/4b5edbfe94a34a91804dfa89900ca350
PageCount 10
ParticipantIDs doaj_primary_oai_doaj_org_article_4b5edbfe94a34a91804dfa89900ca350
crossref_primary_10_4204_EPTCS_306_51
PublicationCentury 2000
PublicationDate 2019-09-19
PublicationDateYYYYMMDD 2019-09-19
PublicationDate_xml – month: 09
  year: 2019
  text: 2019-09-19
  day: 19
PublicationDecade 2010
PublicationTitle Electronic proceedings in theoretical computer science
PublicationYear 2019
Publisher Open Publishing Association
Publisher_xml – name: Open Publishing Association
References xai
Voigt (gdpr) 2017
Zeng (quickfoil) 2014; 8
Riguzzi (alephswiprolog) 2016
Gelfond (gelfond-book) 2014
Shakerin (AAAI2018) 2019
Srinivasan (aleph) 2001
Ribeiro (lime) 2016
Gan (huim) 2018; 8
McMillan (mcmillan) 2005; 345
Muggleton (ilp20) 2012; 86
Baral (baral) 2003
Sakama (sakama05) 2005; 6
Wachter (counterfactual) 2017; 31
Chen (xgboost) 2016
Muggleton (ilp) 1991; 8
Shakerin (fold) 2017; 17
Quinlan (foil) 1990; 5
References_xml – start-page: 3052
  volume-title: AAAI
  year: 2019
  ident: AAAI2018
  article-title: Induction of Non-Monotonic Logic Programs to Explain Boosted Tree Models Using LIME
  doi: 10.1609/aaai.v33i01.33013052
– volume-title: The EU General Data Protection Regulation (GDPR): A Practical Guide
  year: 2017
  ident: gdpr
  doi: 10.1007/978-3-319-57959-7
– volume: 17
  start-page: 1010
  year: 2017
  ident: fold
  article-title: A new algorithm to automate inductive learning of default theories
  publication-title: TPLP
  doi: 10.1017/S1471068417000333
– volume: 8
  issue: 2
  year: 2018
  ident: huim
  article-title: A survey of incremental high-utility itemset mining
  publication-title: Wiley Interdiscip. Rev. Data Min. Knowl. Discov.
  doi: 10.1609/aaai.v33i01.33013052
– volume-title: Explainable Artificial Intelligence (XAI)
  ident: xai
  doi: 10.23919/MIPRO.2018.8400040
– start-page: 785
  volume-title: Proceedings of the 22Nd ACM SIGKDD
  year: 2016
  ident: xgboost
  article-title: XGBoost: A Scalable Tree Boosting System
  doi: 10.1145/2939672.2939785
– volume-title: The Aleph Manual
  year: 2001
  ident: aleph
– volume: 8
  start-page: 197
  issue: 3
  year: 2014
  ident: quickfoil
  article-title: QuickFOIL: Scalable Inductive Logic Programming
  publication-title: Proc. VLDB Endow.
  doi: 10.14778/2735508.2735510
– volume: 8
  start-page: 295
  issue: 4
  year: 1991
  ident: ilp
  article-title: Inductive Logic Programming
  publication-title: New Gen. Comput.
  doi: 10.1007/BF03037089
– start-page: 1135
  volume-title: Proceedings of the 22nd ACM SIGKDD 2016
  year: 2016
  ident: lime
  article-title: "Why Should I Trust You?": Explaining the Predictions of Any Classifier
  doi: 10.1145/2939672.2939778
– volume-title: Knowledge representation, reasoning and declarative problem solving
  year: 2003
  ident: baral
  doi: 10.1017/CBO9780511543357
– volume: 31
  year: 2017
  ident: counterfactual
  article-title: Counterfactual explanations without opening the black box: Automated decisions and the GDPR.(2017)
  publication-title: Harvard Journal of Law & Technology
  doi: 10.2139/ssrn.3063289
– volume-title: ALEPH in SWI-Prolog
  year: 2016
  ident: alephswiprolog
– volume: 86
  start-page: 3
  issue: 1
  year: 2012
  ident: ilp20
  article-title: ILP Turns 20
  publication-title: Mach. Learn.
  doi: 10.1007/s10994-011-5259-2
– volume: 5
  start-page: 239
  year: 1990
  ident: foil
  article-title: Learning Logical Definitions from Relations
  publication-title: Machine Learning
  doi: 10.1007/BF00117105
– volume: 345
  start-page: 101
  issue: 1
  year: 2005
  ident: mcmillan
  article-title: An Interpolating Theorem Prover
  publication-title: Theor. Comput. Sci.
  doi: 10.1016/j.tcs.2005.07.003
– volume: 6
  start-page: 203
  issue: 2
  year: 2005
  ident: sakama05
  article-title: Induction from answer sets in nonmonotonic logic programs
  publication-title: ACM Trans. Comput. Log.
  doi: 10.1145/1055686.1055687
– volume-title: Knowledge Representation, Reasoning, and the Design of Intelligent Agents: The Answer-Set Programming Approach
  year: 2014
  ident: gelfond-book
  doi: 10.1017/CBO9781139342124
SSID ssj0000392496
Score 2.0999787
Snippet We present a fast and scalable algorithm to induce non-monotonic logic programs from statistical learning models. We reduce the problem of search for best...
SourceID doaj
crossref
SourceType Open Website
Index Database
StartPage 379
Title Induction of Non-monotonic Logic Programs To Explain Statistical Learning Models
URI https://doaj.org/article/4b5edbfe94a34a91804dfa89900ca350
Volume 306
WOSCitedRecordID wos001050029200022&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: PRVAON
  databaseName: DOAJ Directory of Open Access Journals
  customDbUrl:
  eissn: 2075-2180
  dateEnd: 20201231
  omitProxy: false
  ssIdentifier: ssj0000392496
  issn: 2075-2180
  databaseCode: DOA
  dateStart: 20090101
  isFulltext: true
  titleUrlDefault: https://www.doaj.org/
  providerName: Directory of Open Access Journals
– providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  customDbUrl:
  eissn: 2075-2180
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000392496
  issn: 2075-2180
  databaseCode: M~E
  dateStart: 20090101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1NS8QwFAwiHrz4LX6Tgx6jSZs0yVFlxYtLwRW8ldd8eJGt7K4e_e2-pFXWkxcvpZTSlnlt3xuYzBByDhEqXnjJQjDApLeaQSs8sy2P0VdJiWFy2IQej83zs62Xor6SJqy3B-6Bu5KtCr6NwUooJVhhuPQRkCVw7qDs2TrXdolM5X9wmXhFTpbDnsiwj_Fe9S4LLq9G9eT28RKH5UslfvWjJdv-3F_utsjGMBjS6_6BtslKmO6Qze_QBTp8g7ukTmEbeTEC7SIdd1OGL1K3SA63NAUnO1r3kqs5nXQ0SeyQ-9M0U2ZLZrzFYKn6QlMO2ut8jzzdjSa392yIRWCuUGbBrNfOBOUBTLTSc-t5bKMIQvhocVcFZQFJJkiAImhrizYa1xaxLEFz0OU-WZ1203BAaMW9dK6N3lVVWtNqFAhteLBYM43TwSG5-AaneevdLxpkDQnEJoPYIIiNEofkJiH3c07yrM4HsJLNUMnmr0oe_cdFjsk6jjRZBSbsCVldzN7DKVlzHwjy7Cy_JLh9-Bx9Aa6qwss
linkProvider Directory of Open Access Journals
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=Induction+of+Non-monotonic+Logic+Programs+To+Explain+Statistical+Learning+Models&rft.jtitle=Electronic+proceedings+in+theoretical+computer+science&rft.au=Farhad+Shakerin&rft.date=2019-09-19&rft.pub=Open+Publishing+Association&rft.issn=2075-2180&rft.eissn=2075-2180&rft.volume=306&rft.issue=Proc.+ICLP+2019&rft.spage=379&rft.epage=388&rft_id=info:doi/10.4204%2FEPTCS.306.51&rft.externalDBID=DOA&rft.externalDocID=oai_doaj_org_article_4b5edbfe94a34a91804dfa89900ca350
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2075-2180&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2075-2180&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2075-2180&client=summon