Centroid Stabilized Fuzzy Tukey Quartile and Z Curve Neural Network based Outlier Detection
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | INFOCOMP: Jornal de Ciência da Computação |
Texto Completo: | https://infocomp.dcc.ufla.br/index.php/infocomp/article/view/2433 |
Resumo: | This paper presents a new method called, Centroid Stabilized Tukey Quartile and Z-curve Neural Network (CSTQ-ZNN) for outlier detection. The main purpose of this paper is to investigate numerous factors that hide outliers and produce the best clustering results via noise reduction, perpetual outlier identification, and centroid stabilization. Moreover unusual outliers are identified using fuzzy clustering and improve computational efficiency by means of centroid stabilization via Tukey Quartile function. The CSTQ-ZNN method is split into two sections. They are clustering of the data points and outlier detection. First Centroid Stabilized Fuzzy Tukey Quartile-based Clustering model is applied to the raw NIFT-50 Stock Market Dataset. Next, with the processed clusters as input, Deep Z-Curve Neural Network model is presented for outlier detection. In contrast, after an analysis of comprehensive experiments performed to validate the CSTQ-ZNN method via comparisons against existing methods and benchmark performance metrics, we found that our proposed method performs better than existing methods in terms of time complexity, error tolerance, true negative rate and outlier detection accuracy. |
id |
UFLA-5_4e379718025681b5e57a01f3ec18127e |
---|---|
oai_identifier_str |
oai:infocomp.dcc.ufla.br:article/2433 |
network_acronym_str |
UFLA-5 |
network_name_str |
INFOCOMP: Jornal de Ciência da Computação |
repository_id_str |
|
spelling |
Centroid Stabilized Fuzzy Tukey Quartile and Z Curve Neural Network based Outlier DetectionThis paper presents a new method called, Centroid Stabilized Tukey Quartile and Z-curve Neural Network (CSTQ-ZNN) for outlier detection. The main purpose of this paper is to investigate numerous factors that hide outliers and produce the best clustering results via noise reduction, perpetual outlier identification, and centroid stabilization. Moreover unusual outliers are identified using fuzzy clustering and improve computational efficiency by means of centroid stabilization via Tukey Quartile function. The CSTQ-ZNN method is split into two sections. They are clustering of the data points and outlier detection. First Centroid Stabilized Fuzzy Tukey Quartile-based Clustering model is applied to the raw NIFT-50 Stock Market Dataset. Next, with the processed clusters as input, Deep Z-Curve Neural Network model is presented for outlier detection. In contrast, after an analysis of comprehensive experiments performed to validate the CSTQ-ZNN method via comparisons against existing methods and benchmark performance metrics, we found that our proposed method performs better than existing methods in terms of time complexity, error tolerance, true negative rate and outlier detection accuracy.Editora da UFLA2022-12-19info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://infocomp.dcc.ufla.br/index.php/infocomp/article/view/2433INFOCOMP Journal of Computer Science; Vol. 21 No. 2 (2022): December 20221982-33631807-4545reponame:INFOCOMP: Jornal de Ciência da Computaçãoinstname:Universidade Federal de Lavras (UFLA)instacron:UFLAenghttps://infocomp.dcc.ufla.br/index.php/infocomp/article/view/2433/584Copyright (c) 2022 S RAJALAKSHMI, P MADHUBALAinfo:eu-repo/semantics/openAccessS RAJALAKSHMIP MADHUBALA2022-12-19T14:48:47Zoai:infocomp.dcc.ufla.br:article/2433Revistahttps://infocomp.dcc.ufla.br/index.php/infocompPUBhttps://infocomp.dcc.ufla.br/index.php/infocomp/oaiinfocomp@dcc.ufla.br||apfreire@dcc.ufla.br1982-33631807-4545opendoar:2024-05-21T19:54:48.142305INFOCOMP: Jornal de Ciência da Computação - Universidade Federal de Lavras (UFLA)true |
dc.title.none.fl_str_mv |
Centroid Stabilized Fuzzy Tukey Quartile and Z Curve Neural Network based Outlier Detection |
title |
Centroid Stabilized Fuzzy Tukey Quartile and Z Curve Neural Network based Outlier Detection |
spellingShingle |
Centroid Stabilized Fuzzy Tukey Quartile and Z Curve Neural Network based Outlier Detection S RAJALAKSHMI |
title_short |
Centroid Stabilized Fuzzy Tukey Quartile and Z Curve Neural Network based Outlier Detection |
title_full |
Centroid Stabilized Fuzzy Tukey Quartile and Z Curve Neural Network based Outlier Detection |
title_fullStr |
Centroid Stabilized Fuzzy Tukey Quartile and Z Curve Neural Network based Outlier Detection |
title_full_unstemmed |
Centroid Stabilized Fuzzy Tukey Quartile and Z Curve Neural Network based Outlier Detection |
title_sort |
Centroid Stabilized Fuzzy Tukey Quartile and Z Curve Neural Network based Outlier Detection |
author |
S RAJALAKSHMI |
author_facet |
S RAJALAKSHMI P MADHUBALA |
author_role |
author |
author2 |
P MADHUBALA |
author2_role |
author |
dc.contributor.author.fl_str_mv |
S RAJALAKSHMI P MADHUBALA |
description |
This paper presents a new method called, Centroid Stabilized Tukey Quartile and Z-curve Neural Network (CSTQ-ZNN) for outlier detection. The main purpose of this paper is to investigate numerous factors that hide outliers and produce the best clustering results via noise reduction, perpetual outlier identification, and centroid stabilization. Moreover unusual outliers are identified using fuzzy clustering and improve computational efficiency by means of centroid stabilization via Tukey Quartile function. The CSTQ-ZNN method is split into two sections. They are clustering of the data points and outlier detection. First Centroid Stabilized Fuzzy Tukey Quartile-based Clustering model is applied to the raw NIFT-50 Stock Market Dataset. Next, with the processed clusters as input, Deep Z-Curve Neural Network model is presented for outlier detection. In contrast, after an analysis of comprehensive experiments performed to validate the CSTQ-ZNN method via comparisons against existing methods and benchmark performance metrics, we found that our proposed method performs better than existing methods in terms of time complexity, error tolerance, true negative rate and outlier detection accuracy. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-12-19 |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://infocomp.dcc.ufla.br/index.php/infocomp/article/view/2433 |
url |
https://infocomp.dcc.ufla.br/index.php/infocomp/article/view/2433 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
https://infocomp.dcc.ufla.br/index.php/infocomp/article/view/2433/584 |
dc.rights.driver.fl_str_mv |
Copyright (c) 2022 S RAJALAKSHMI, P MADHUBALA info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Copyright (c) 2022 S RAJALAKSHMI, P MADHUBALA |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Editora da UFLA |
publisher.none.fl_str_mv |
Editora da UFLA |
dc.source.none.fl_str_mv |
INFOCOMP Journal of Computer Science; Vol. 21 No. 2 (2022): December 2022 1982-3363 1807-4545 reponame:INFOCOMP: Jornal de Ciência da Computação instname:Universidade Federal de Lavras (UFLA) instacron:UFLA |
instname_str |
Universidade Federal de Lavras (UFLA) |
instacron_str |
UFLA |
institution |
UFLA |
reponame_str |
INFOCOMP: Jornal de Ciência da Computação |
collection |
INFOCOMP: Jornal de Ciência da Computação |
repository.name.fl_str_mv |
INFOCOMP: Jornal de Ciência da Computação - Universidade Federal de Lavras (UFLA) |
repository.mail.fl_str_mv |
infocomp@dcc.ufla.br||apfreire@dcc.ufla.br |
_version_ |
1799874742694445056 |