Academic Journal
Integrating Structural Causal Model Ontologies with LIME for Fair Machine Learning Explanations in Educational Admissions
| العنوان: | Integrating Structural Causal Model Ontologies with LIME for Fair Machine Learning Explanations in Educational Admissions |
|---|---|
| المؤلفون: | Igoche, Bern Igoche, Matthew, Olumuyiwa, Bednar, Peter, Gegov, Alexander |
| المصدر: | Journal of Computing Theories and Applications; Vol. 2 No. 1 (2024): JCTA 2(1) 2024; 65-85 ; 3024-9104 |
| بيانات النشر: | Universitas Dian Nuswantoro |
| سنة النشر: | 2024 |
| المجموعة: | Ruang Publikasi Ilmiah Universitas Dian Nuswantoro |
| مصطلحات موضوعية: | knowledge discovery in databases (KDD), structural causal model (SCM), ontology, local interpretable model-agnostic explanations (LIME), fairness |
| الوصف: | This study employed knowledge discovery in databases (KDD) to extract and discover knowledge from the Benue State Polytechnic (Benpoly) admission database and used a structural causal model (SCM) ontological framework to represent the admission process in the Nigerian polytechnic education system. The SCM ontology identified important causal relations in features needed to model the admission process and was validated using the conditional independence test (CIT) criteria. The SCM ontology was further employed to identify and constrain input features causing bias in the local interpretable model-agnostic explanations (LIME) framework applied to machine learning (ML) black-box predictions. The ablation process produced more stable LIME explanations devoid of fairness bias compared to LIME without ablation, with higher prediction accuracy (91% vs. 89%) and F1 scores (95% vs. 94%). The study also compared the performance of different ML models, including Gaussian Naïve Bayes, Decision Trees, and Logistic Regression, before and after ablation. The limitation is that the SCM ontology is qualitative and context-specific, so the fair-LIME framework can only be extrapolated to similar contexts. Future work could compare other explanation frameworks like Shapley on the same dataset. Overall, this study demonstrates a novel approach to enforcing fairness in ML explanations by integrating qualitative SCM ontologies with quantitative ML/LIME methods. |
| نوع الوثيقة: | article in journal/newspaper |
| وصف الملف: | application/pdf |
| اللغة: | English |
| Relation: | https://publikasi.dinus.ac.id/index.php/jcta/article/view/10501/4542; https://publikasi.dinus.ac.id/index.php/jcta/article/view/10501 |
| DOI: | 10.62411/jcta.10501 |
| الاتاحة: | https://publikasi.dinus.ac.id/index.php/jcta/article/view/10501 https://doi.org/10.62411/jcta.10501 |
| Rights: | Copyright (c) 2024 Bern Igoche Igoche ; https://creativecommons.org/licenses/by/4.0 |
| رقم الانضمام: | edsbas.F8116F7B |
| قاعدة البيانات: | BASE |
| ResultId |
1 |
|---|---|
| Header |
edsbas BASE edsbas.F8116F7B 995 3 Academic Journal academicJournal 995.377258300781 |
| PLink |
https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&scope=site&db=edsbas&AN=edsbas.F8116F7B&custid=s6537998&authtype=sso |
| FullText |
Array
(
[Availability] => 0
)
Array ( [0] => Array ( [Url] => https://publikasi.dinus.ac.id/index.php/jcta/article/view/10501# [Name] => EDS - BASE [Category] => fullText [Text] => View record in BASE [MouseOverText] => View record in BASE ) ) |
| Items |
Array
(
[Name] => Title
[Label] => Title
[Group] => Ti
[Data] => Integrating Structural Causal Model Ontologies with LIME for Fair Machine Learning Explanations in Educational Admissions
)
Array ( [Name] => Author [Label] => Authors [Group] => Au [Data] => <searchLink fieldCode="AR" term="%22Igoche%2C+Bern+Igoche%22">Igoche, Bern Igoche</searchLink><br /><searchLink fieldCode="AR" term="%22Matthew%2C+Olumuyiwa%22">Matthew, Olumuyiwa</searchLink><br /><searchLink fieldCode="AR" term="%22Bednar%2C+Peter%22">Bednar, Peter</searchLink><br /><searchLink fieldCode="AR" term="%22Gegov%2C+Alexander%22">Gegov, Alexander</searchLink> ) Array ( [Name] => TitleSource [Label] => Source [Group] => Src [Data] => Journal of Computing Theories and Applications; Vol. 2 No. 1 (2024): JCTA 2(1) 2024; 65-85 ; 3024-9104 ) Array ( [Name] => Publisher [Label] => Publisher Information [Group] => PubInfo [Data] => Universitas Dian Nuswantoro ) Array ( [Name] => DatePubCY [Label] => Publication Year [Group] => Date [Data] => 2024 ) Array ( [Name] => Subset [Label] => Collection [Group] => HoldingsInfo [Data] => Ruang Publikasi Ilmiah Universitas Dian Nuswantoro ) Array ( [Name] => Subject [Label] => Subject Terms [Group] => Su [Data] => <searchLink fieldCode="DE" term="%22knowledge+discovery+in+databases+%28KDD%29%22">knowledge discovery in databases (KDD)</searchLink><br /><searchLink fieldCode="DE" term="%22structural+causal+model+%28SCM%29%22">structural causal model (SCM)</searchLink><br /><searchLink fieldCode="DE" term="%22ontology%22">ontology</searchLink><br /><searchLink fieldCode="DE" term="%22local+interpretable+model-agnostic+explanations+%28LIME%29%22">local interpretable model-agnostic explanations (LIME)</searchLink><br /><searchLink fieldCode="DE" term="%22fairness%22">fairness</searchLink> ) Array ( [Name] => Abstract [Label] => Description [Group] => Ab [Data] => This study employed knowledge discovery in databases (KDD) to extract and discover knowledge from the Benue State Polytechnic (Benpoly) admission database and used a structural causal model (SCM) ontological framework to represent the admission process in the Nigerian polytechnic education system. The SCM ontology identified important causal relations in features needed to model the admission process and was validated using the conditional independence test (CIT) criteria. The SCM ontology was further employed to identify and constrain input features causing bias in the local interpretable model-agnostic explanations (LIME) framework applied to machine learning (ML) black-box predictions. The ablation process produced more stable LIME explanations devoid of fairness bias compared to LIME without ablation, with higher prediction accuracy (91% vs. 89%) and F1 scores (95% vs. 94%). The study also compared the performance of different ML models, including Gaussian Naïve Bayes, Decision Trees, and Logistic Regression, before and after ablation. The limitation is that the SCM ontology is qualitative and context-specific, so the fair-LIME framework can only be extrapolated to similar contexts. Future work could compare other explanation frameworks like Shapley on the same dataset. Overall, this study demonstrates a novel approach to enforcing fairness in ML explanations by integrating qualitative SCM ontologies with quantitative ML/LIME methods. ) Array ( [Name] => TypeDocument [Label] => Document Type [Group] => TypDoc [Data] => article in journal/newspaper ) Array ( [Name] => Format [Label] => File Description [Group] => SrcInfo [Data] => application/pdf ) Array ( [Name] => Language [Label] => Language [Group] => Lang [Data] => English ) Array ( [Name] => NoteTitleSource [Label] => Relation [Group] => SrcInfo [Data] => https://publikasi.dinus.ac.id/index.php/jcta/article/view/10501/4542; https://publikasi.dinus.ac.id/index.php/jcta/article/view/10501 ) Array ( [Name] => DOI [Label] => DOI [Group] => ID [Data] => 10.62411/jcta.10501 ) Array ( [Name] => URL [Label] => Availability [Group] => URL [Data] => https://publikasi.dinus.ac.id/index.php/jcta/article/view/10501<br />https://doi.org/10.62411/jcta.10501 ) Array ( [Name] => Copyright [Label] => Rights [Group] => Cpyrght [Data] => Copyright (c) 2024 Bern Igoche Igoche ; https://creativecommons.org/licenses/by/4.0 ) Array ( [Name] => AN [Label] => Accession Number [Group] => ID [Data] => edsbas.F8116F7B ) |
| RecordInfo |
Array
(
[BibEntity] => Array
(
[Identifiers] => Array
(
[0] => Array
(
[Type] => doi
[Value] => 10.62411/jcta.10501
)
)
[Languages] => Array
(
[0] => Array
(
[Text] => English
)
)
[Subjects] => Array
(
[0] => Array
(
[SubjectFull] => knowledge discovery in databases (KDD)
[Type] => general
)
[1] => Array
(
[SubjectFull] => structural causal model (SCM)
[Type] => general
)
[2] => Array
(
[SubjectFull] => ontology
[Type] => general
)
[3] => Array
(
[SubjectFull] => local interpretable model-agnostic explanations (LIME)
[Type] => general
)
[4] => Array
(
[SubjectFull] => fairness
[Type] => general
)
)
[Titles] => Array
(
[0] => Array
(
[TitleFull] => Integrating Structural Causal Model Ontologies with LIME for Fair Machine Learning Explanations in Educational Admissions
[Type] => main
)
)
)
[BibRelationships] => Array
(
[HasContributorRelationships] => Array
(
[0] => Array
(
[PersonEntity] => Array
(
[Name] => Array
(
[NameFull] => Igoche, Bern Igoche
)
)
)
[1] => Array
(
[PersonEntity] => Array
(
[Name] => Array
(
[NameFull] => Matthew, Olumuyiwa
)
)
)
[2] => Array
(
[PersonEntity] => Array
(
[Name] => Array
(
[NameFull] => Bednar, Peter
)
)
)
[3] => Array
(
[PersonEntity] => Array
(
[Name] => Array
(
[NameFull] => Gegov, Alexander
)
)
)
)
[IsPartOfRelationships] => Array
(
[0] => Array
(
[BibEntity] => Array
(
[Dates] => Array
(
[0] => Array
(
[D] => 01
[M] => 01
[Type] => published
[Y] => 2024
)
)
[Identifiers] => Array
(
[0] => Array
(
[Type] => issn-locals
[Value] => edsbas
)
[1] => Array
(
[Type] => issn-locals
[Value] => edsbas.oa
)
)
[Titles] => Array
(
[0] => Array
(
[TitleFull] => Journal of Computing Theories and Applications; Vol. 2 No. 1 (2024): JCTA 2(1) 2024
[Type] => main
)
)
)
)
)
)
)
|
| IllustrationInfo |