Publicação

Data mining in retail sector

Ver documento

Detalhes bibliográficos
Resumo:The retailsectorisoneofPortugal’smostrelevanteconomicactivitiesbecausein2021it was the sector that employed the most Portuguese people and the second largest contributor t gross fixed capital formation. Despite this,in the same year it was the third sector with th most accidents at work.There fore, this master’s thesis aims to apply data mining techniques to improve work accidents prevention using internal and external data from a Portuguese retail company. The company provide dinternal data on stores, accidents and employees, which was the nintegrated with weather information collected via anexternal API. Th correlation analysis was applied separating the data by store and by district and idemonstrated a weak correlation between the features studied and the occurrence of accidents at work. Further more, ML models were trained using the same features with the intention of classifying the data between occurrence(1) ornon-occurrence(0) ofaccidents, also separating by store and by district while comparing 8ML algorithms. Another categorization of stores was testedusing a clustering algorithm along with a number of clusters optimizing method.The stores were then dividedin to clusters so that the same correlation analysis and ML classification models could be implemented for comparison. The correlation analysis per-cluster yielded no different results from the previous ones. On the other hand, the classificationa lgorithms trained by cluster performed better,with the Multilayer Perceptron algorithm obtaining Recall = 0.7959.
Autores principais:Borges, Lucas D.
Assunto:Data mining Retail Work accidents
Ano:2024
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso restrito
Instituição associada:Instituto Politécnico de Bragança
Idioma:inglês
Origem:Biblioteca Digital do IPB
_version_ 1867173001193062400
author Borges, Lucas D.
author_facet Borges, Lucas D.
author_role author
contributor_name_str_mv Pereira, Ana I.
Vaz, Clara B.
Lanes, Matusalém M.
Biblioteca Digital do IPB
country_str PT
creators_json_txt [{\"Person.name\":\"Borges, Lucas D.\"}]
datacite.contributors.contributor.contributorName.fl_str_mv Pereira, Ana I.
Vaz, Clara B.
Lanes, Matusalém M.
Biblioteca Digital do IPB
datacite.creators.creator.creatorName.fl_str_mv Borges, Lucas D.
datacite.date.Accepted.fl_str_mv 2024-01-01T00:00:00Z
datacite.date.available.fl_str_mv 2024-10-01T11:28:29Z
datacite.date.embargoed.fl_str_mv 2024-10-01T11:28:29Z
datacite.rights.fl_str_mv http://purl.org/coar/access_right/c_16ec
datacite.subjects.subject.fl_str_mv Data mining
Retail
Work accidents
datacite.titles.title.fl_str_mv Data mining in retail sector
dc.contributor.none.fl_str_mv Pereira, Ana I.
Vaz, Clara B.
Lanes, Matusalém M.
Biblioteca Digital do IPB
dc.creator.none.fl_str_mv Borges, Lucas D.
dc.date.Accepted.fl_str_mv 2024-01-01T00:00:00Z
dc.date.available.fl_str_mv 2024-10-01T11:28:29Z
dc.date.embargoed.fl_str_mv 2024-10-01T11:28:29Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv http://hdl.handle.net/10198/30300
dc.language.none.fl_str_mv eng
dc.rights.cclincense.fl_str_mv http://creativecommons.org/licenses/by-nc/4.0/
dc.rights.none.fl_str_mv http://purl.org/coar/access_right/c_16ec
dc.subject.none.fl_str_mv Data mining
Retail
Work accidents
dc.title.fl_str_mv Data mining in retail sector
dc.type.none.fl_str_mv http://purl.org/coar/resource_type/c_bdcc
description The retailsectorisoneofPortugal’smostrelevanteconomicactivitiesbecausein2021it was the sector that employed the most Portuguese people and the second largest contributor t gross fixed capital formation. Despite this,in the same year it was the third sector with th most accidents at work.There fore, this master’s thesis aims to apply data mining techniques to improve work accidents prevention using internal and external data from a Portuguese retail company. The company provide dinternal data on stores, accidents and employees, which was the nintegrated with weather information collected via anexternal API. Th correlation analysis was applied separating the data by store and by district and idemonstrated a weak correlation between the features studied and the occurrence of accidents at work. Further more, ML models were trained using the same features with the intention of classifying the data between occurrence(1) ornon-occurrence(0) ofaccidents, also separating by store and by district while comparing 8ML algorithms. Another categorization of stores was testedusing a clustering algorithm along with a number of clusters optimizing method.The stores were then dividedin to clusters so that the same correlation analysis and ML classification models could be implemented for comparison. The correlation analysis per-cluster yielded no different results from the previous ones. On the other hand, the classificationa lgorithms trained by cluster performed better,with the Multilayer Perceptron algorithm obtaining Recall = 0.7959.
dirty 0
eu_rights_str_mv restrictedAccess
format masterThesis
fulltext.url.fl_str_mv https://bibliotecadigital.ipb.pt/bitstreams/2e17e9e6-e002-487e-ac53-cb7d43a40cbf/download
id ipb_52230683df75475fffaecad0ab4d5188
identifier.url.fl_str_mv http://hdl.handle.net/10198/30300
instacron_str ipb
institution Instituto Politécnico de Bragança
instname_str Instituto Politécnico de Bragança
language eng
network_acronym_str ipb
network_name_str Biblioteca Digital do IPB
oai_identifier_str oai:bibliotecadigital.ipb.pt:10198/30300
organization_str_mv urn:organizationAcronym:ipb
person_str_mv Borges, Lucas D.
publishDate 2024
reponame_str Biblioteca Digital do IPB
repository_id_str urn:repositoryAcronym:ipb
service_str_mv urn:repositoryAcronym:ipb
spelling engpt_PTThe retailsectorisoneofPortugal’smostrelevanteconomicactivitiesbecausein2021it was the sector that employed the most Portuguese people and the second largest contributor t gross fixed capital formation. Despite this,in the same year it was the third sector with th most accidents at work.There fore, this master’s thesis aims to apply data mining techniques to improve work accidents prevention using internal and external data from a Portuguese retail company. The company provide dinternal data on stores, accidents and employees, which was the nintegrated with weather information collected via anexternal API. Th correlation analysis was applied separating the data by store and by district and idemonstrated a weak correlation between the features studied and the occurrence of accidents at work. Further more, ML models were trained using the same features with the intention of classifying the data between occurrence(1) ornon-occurrence(0) ofaccidents, also separating by store and by district while comparing 8ML algorithms. Another categorization of stores was testedusing a clustering algorithm along with a number of clusters optimizing method.The stores were then dividedin to clusters so that the same correlation analysis and ML classification models could be implemented for comparison. The correlation analysis per-cluster yielded no different results from the previous ones. On the other hand, the classificationa lgorithms trained by cluster performed better,with the Multilayer Perceptron algorithm obtaining Recall = 0.7959.application/pdfpt_PTData mining in retail sectorBorges, Lucas D.Pereira, Ana I.Vaz, Clara B.Lanes, Matusalém M.HostingInstitutionOrganizationalBiblioteca Digital do IPBe-mailmailto:dspace@ipb.ptdspace@ipb.ptURNurn:tid:2036983392024-10-01T11:28:29Z20242024-01-01T00:00:00ZHandlehttp://hdl.handle.net/10198/30300http://purl.org/coar/access_right/c_16ecrestricted accessData miningRetailWork accidents6402495 bytesliteraturehttp://purl.org/coar/resource_type/c_bdccmaster thesis2024http://creativecommons.org/licenses/by-nc/4.0/http://purl.org/coar/access_right/c_16ecapplication/pdffulltexthttps://bibliotecadigital.ipb.pt/bitstreams/2e17e9e6-e002-487e-ac53-cb7d43a40cbf/download
spellingShingle Data mining in retail sector
Borges, Lucas D.
Data mining
Retail
Work accidents
status SINGLETON
subject.fl_str_mv Data mining
Retail
Work accidents
title Data mining in retail sector
title_full Data mining in retail sector
title_fullStr Data mining in retail sector
title_full_unstemmed Data mining in retail sector
title_short Data mining in retail sector
title_sort Data mining in retail sector
topic Data mining
Retail
Work accidents
topic_facet Data mining
Retail
Work accidents
url http://hdl.handle.net/10198/30300
visible 1