Apriori-Roaring-Parallel: Frequent pattern mining based on compressed bitmaps with OpenMP

Detalhes bibliográficos
Autor(a) principal: Colombo, Alexandre [UNESP]
Data de Publicação: 2021
Outros Autores: Spolon, Roberta [UNESP], Lobato, Renata Spolon [UNESP], Manacero, Aleardo [UNESP], Cavenaghi, Marcos Antonio
Tipo de documento: Artigo de conferência
Idioma: eng
Título da fonte: Repositório Institucional da UNESP
Texto Completo: http://dx.doi.org/10.1109/ISCC53001.2021.9631495
http://hdl.handle.net/11449/230253
Resumo: Mining association rules is a process which consists in extracting knowledge from datasets. This is a widely used technique to analyze customer purchasing patterns, and its process is segmented in two main phases: mining frequent sets and formulating association rules. Several approaches were developed for the first phase of the mining process whose main objective was to reduce execution time. However, as all available datasets are very large (Big Data), there is a limitation regarding its application in these new sets due to excessive memory usage. We propose the Apriori-Roaring-Parallel which explores parallelism in shared memory and demands less memory usage during the mining process. In order to achieve this memory usage reduction, the Apriori-Roaring-Parallel method employs compressed bitmap structures to represent the datasets. The results obtained show that the Apriori-Roaring-Parallel method uses memory efficiently when compared to other methods.
id UNSP_aae8d6b688bd7083bcc7461aeb977aa5
oai_identifier_str oai:repositorio.unesp.br:11449/230253
network_acronym_str UNSP
network_name_str Repositório Institucional da UNESP
repository_id_str 2946
spelling Apriori-Roaring-Parallel: Frequent pattern mining based on compressed bitmaps with OpenMPAssociation RulesBitmap CompressionData MiningIdentification of Frequent SetsMining association rules is a process which consists in extracting knowledge from datasets. This is a widely used technique to analyze customer purchasing patterns, and its process is segmented in two main phases: mining frequent sets and formulating association rules. Several approaches were developed for the first phase of the mining process whose main objective was to reduce execution time. However, as all available datasets are very large (Big Data), there is a limitation regarding its application in these new sets due to excessive memory usage. We propose the Apriori-Roaring-Parallel which explores parallelism in shared memory and demands less memory usage during the mining process. In order to achieve this memory usage reduction, the Apriori-Roaring-Parallel method employs compressed bitmap structures to represent the datasets. The results obtained show that the Apriori-Roaring-Parallel method uses memory efficiently when compared to other methods.São Paulo State University UNESP Computing DepartmentSão Paulo State University UNESP Department of Computer Science And Statistics, São José do Rio PretoHumber Institute of Technology And Advanced Learning Faculty of BusinessSão Paulo State University UNESP Computing DepartmentSão Paulo State University UNESP Department of Computer Science And Statistics, São José do Rio PretoUniversidade Estadual Paulista (UNESP)Faculty of BusinessColombo, Alexandre [UNESP]Spolon, Roberta [UNESP]Lobato, Renata Spolon [UNESP]Manacero, Aleardo [UNESP]Cavenaghi, Marcos Antonio2022-04-29T08:38:44Z2022-04-29T08:38:44Z2021-01-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/conferenceObjecthttp://dx.doi.org/10.1109/ISCC53001.2021.9631495Proceedings - IEEE Symposium on Computers and Communications, v. 2021-September.1530-1346http://hdl.handle.net/11449/23025310.1109/ISCC53001.2021.96314952-s2.0-85123186952Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengProceedings - IEEE Symposium on Computers and Communicationsinfo:eu-repo/semantics/openAccess2022-04-29T08:38:44Zoai:repositorio.unesp.br:11449/230253Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T23:03:00.359995Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false
dc.title.none.fl_str_mv Apriori-Roaring-Parallel: Frequent pattern mining based on compressed bitmaps with OpenMP
title Apriori-Roaring-Parallel: Frequent pattern mining based on compressed bitmaps with OpenMP
spellingShingle Apriori-Roaring-Parallel: Frequent pattern mining based on compressed bitmaps with OpenMP
Colombo, Alexandre [UNESP]
Association Rules
Bitmap Compression
Data Mining
Identification of Frequent Sets
title_short Apriori-Roaring-Parallel: Frequent pattern mining based on compressed bitmaps with OpenMP
title_full Apriori-Roaring-Parallel: Frequent pattern mining based on compressed bitmaps with OpenMP
title_fullStr Apriori-Roaring-Parallel: Frequent pattern mining based on compressed bitmaps with OpenMP
title_full_unstemmed Apriori-Roaring-Parallel: Frequent pattern mining based on compressed bitmaps with OpenMP
title_sort Apriori-Roaring-Parallel: Frequent pattern mining based on compressed bitmaps with OpenMP
author Colombo, Alexandre [UNESP]
author_facet Colombo, Alexandre [UNESP]
Spolon, Roberta [UNESP]
Lobato, Renata Spolon [UNESP]
Manacero, Aleardo [UNESP]
Cavenaghi, Marcos Antonio
author_role author
author2 Spolon, Roberta [UNESP]
Lobato, Renata Spolon [UNESP]
Manacero, Aleardo [UNESP]
Cavenaghi, Marcos Antonio
author2_role author
author
author
author
dc.contributor.none.fl_str_mv Universidade Estadual Paulista (UNESP)
Faculty of Business
dc.contributor.author.fl_str_mv Colombo, Alexandre [UNESP]
Spolon, Roberta [UNESP]
Lobato, Renata Spolon [UNESP]
Manacero, Aleardo [UNESP]
Cavenaghi, Marcos Antonio
dc.subject.por.fl_str_mv Association Rules
Bitmap Compression
Data Mining
Identification of Frequent Sets
topic Association Rules
Bitmap Compression
Data Mining
Identification of Frequent Sets
description Mining association rules is a process which consists in extracting knowledge from datasets. This is a widely used technique to analyze customer purchasing patterns, and its process is segmented in two main phases: mining frequent sets and formulating association rules. Several approaches were developed for the first phase of the mining process whose main objective was to reduce execution time. However, as all available datasets are very large (Big Data), there is a limitation regarding its application in these new sets due to excessive memory usage. We propose the Apriori-Roaring-Parallel which explores parallelism in shared memory and demands less memory usage during the mining process. In order to achieve this memory usage reduction, the Apriori-Roaring-Parallel method employs compressed bitmap structures to represent the datasets. The results obtained show that the Apriori-Roaring-Parallel method uses memory efficiently when compared to other methods.
publishDate 2021
dc.date.none.fl_str_mv 2021-01-01
2022-04-29T08:38:44Z
2022-04-29T08:38:44Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/conferenceObject
format conferenceObject
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://dx.doi.org/10.1109/ISCC53001.2021.9631495
Proceedings - IEEE Symposium on Computers and Communications, v. 2021-September.
1530-1346
http://hdl.handle.net/11449/230253
10.1109/ISCC53001.2021.9631495
2-s2.0-85123186952
url http://dx.doi.org/10.1109/ISCC53001.2021.9631495
http://hdl.handle.net/11449/230253
identifier_str_mv Proceedings - IEEE Symposium on Computers and Communications, v. 2021-September.
1530-1346
10.1109/ISCC53001.2021.9631495
2-s2.0-85123186952
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv Proceedings - IEEE Symposium on Computers and Communications
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.source.none.fl_str_mv Scopus
reponame:Repositório Institucional da UNESP
instname:Universidade Estadual Paulista (UNESP)
instacron:UNESP
instname_str Universidade Estadual Paulista (UNESP)
instacron_str UNESP
institution UNESP
reponame_str Repositório Institucional da UNESP
collection Repositório Institucional da UNESP
repository.name.fl_str_mv Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)
repository.mail.fl_str_mv
_version_ 1808129484650446848