Seleção de características aplicado ao keystroke dynamics em dispositivos móveis

Detalhes bibliográficos
Autor(a) principal: LEITE, Urbanno Pereira de Siqueira
Data de Publicação: 2018
Tipo de documento: Dissertação
Idioma: por
Título da fonte: Biblioteca Digital de Teses e Dissertações da UFRPE
Texto Completo: http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/7861
Resumo: In this paper, feature selection models are developed for keystroke dynamics on mobile devices. Two models are elaborated, one based on Genetic Algorithm (GA) and another on PSO. The proposed selectors are applied to a public database of keystroke dynamics built from mobile devices. Both methods of selection are used in conjunction with various classifiers: Naive Bayes, Bayes Net, C4.5 (decision tree), Random Forest, k-NN, Support Vector Machine (SVM) and Multilayer Perceptron (MLP). The feature selection methods developed here are evaluated by accuracy, false positive rate (FAR), false negative rate (FRR) - all these measures are obtained from the classifications - and also by the reduction rates of characteristics. The results obtained from the execution of several experiments show that the proposed models were able to add improvements to the measures of performance - when compared to the results of the classifications without selection -, besides reaching high levels of reduction of characteristics. Through a comparative analysis it was also possible to verify that the models developed in this work have performances compatible with other selectors already available in the literature. The proposed methods also call attention for the stability of their behavior, in such a way that the results generated by them have low indices of variability. In this work it was possible to identify the most selected features and also those less chosen by the models, showing that an attribute can be quite selected for a particular classification method, but not so chosen for another classifier. Already analyzing the frequency of selection of characteristics according to their type, it was verified that the two characteristics most selected by the proposed selection methods are attributes inherent to mobile devices.
id URPE_bdd72916a80f82352bf1eb038fc5380c
oai_identifier_str oai:tede2:tede2/7861
network_acronym_str URPE
network_name_str Biblioteca Digital de Teses e Dissertações da UFRPE
repository_id_str
spelling FERREIRA, Tiago Alessandro EspínolaCAVALCANTI, George Darmiton da CunhaGARROZI, CíceroMIRANDA, Péricles Barbosa Cunha dehttp://lattes.cnpq.br/3544721962219698LEITE, Urbanno Pereira de Siqueira2019-02-22T15:24:50Z2018-02-26LEITE, Urbanno Pereira de Siqueira. Seleção de características aplicado ao keystroke dynamics em dispositivos móveis. 2018. 109 f. Dissertação (Programa de Pós-Graduação em Informática Aplicada) - Universidade Federal Rural de Pernambuco, Recife.http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/7861In this paper, feature selection models are developed for keystroke dynamics on mobile devices. Two models are elaborated, one based on Genetic Algorithm (GA) and another on PSO. The proposed selectors are applied to a public database of keystroke dynamics built from mobile devices. Both methods of selection are used in conjunction with various classifiers: Naive Bayes, Bayes Net, C4.5 (decision tree), Random Forest, k-NN, Support Vector Machine (SVM) and Multilayer Perceptron (MLP). The feature selection methods developed here are evaluated by accuracy, false positive rate (FAR), false negative rate (FRR) - all these measures are obtained from the classifications - and also by the reduction rates of characteristics. The results obtained from the execution of several experiments show that the proposed models were able to add improvements to the measures of performance - when compared to the results of the classifications without selection -, besides reaching high levels of reduction of characteristics. Through a comparative analysis it was also possible to verify that the models developed in this work have performances compatible with other selectors already available in the literature. The proposed methods also call attention for the stability of their behavior, in such a way that the results generated by them have low indices of variability. In this work it was possible to identify the most selected features and also those less chosen by the models, showing that an attribute can be quite selected for a particular classification method, but not so chosen for another classifier. Already analyzing the frequency of selection of characteristics according to their type, it was verified that the two characteristics most selected by the proposed selection methods are attributes inherent to mobile devices.Nesse trabalho são desenvolvidos modelos de seleção de características para o keystroke dynamics em dispositivos móveis. Dois modelos são elaborados, um baseado em Algoritmo Genético (GA) e outro em PSO. Os seletores propostos são aplicados a uma base de dados pública do keystroke dynamics construída a partir de dispositivos móveis. Ambos os métodos de seleção são utilizados em conjunto com variados classificadores, são eles: Naive Bayes, Bayes Net, C4.5 (árvore de decisão), Random Forest, k-NN, Support Vector Machine (SVM) e Multilayer Perceptron (MLP). Os métodos de seleção de características aqui desenvolvidos são avaliados a partir das métricas de taxa de acuracidade, falso positivo (FAR), falso negativo (FRR) - todas essas medidas são obtidas a partir das classificações - e também por meio das taxas de redução de características. Os resultados obtidos a partir da execução de vários experimentos mostram que os modelos propostos foram capazes de agregar melhorias às medidas de desempenho - quando comparados aos resultados das classificações sem seleção -, além de alcançarem altos níveis de redução de características. Através de uma análise comparativa foi possível também verificar que os modelos desenvolvidos nesse trabalho possuem desempenhos compatíveis com outros seletores já disponíveis na literatura. Os métodos propostos também chamam atenção pela estabilidade do seu comportamento, de tal forma que os resultados por eles gerados possuem baixos índices de variabilidade. Nesse trabalho foi possível ainda se identificar as características mais selecionadas e também aquelas menos escolhidas pelos modelos, sendo mostrado que um atributo pode ser bastante selecionado para um determinado método de classificação, porém não ser tão escolhido para um outro classificador. Já analisando a frequência de seleção da características de acordo com o seu tipo, verificou-se que as duas características mais selecionadas pelos métodos de seleção propostos são atributos inerentes aos dispositivos móveis.Submitted by Mario BC (mario@bc.ufrpe.br) on 2019-02-22T15:24:50Z No. of bitstreams: 1 Urbanno Pereira de Siqueira Leite.pdf: 1660810 bytes, checksum: e2e67d450cc32a40bfc934c706068368 (MD5)Made available in DSpace on 2019-02-22T15:24:50Z (GMT). No. of bitstreams: 1 Urbanno Pereira de Siqueira Leite.pdf: 1660810 bytes, checksum: e2e67d450cc32a40bfc934c706068368 (MD5) Previous issue date: 2018-02-26application/pdfporUniversidade Federal Rural de PernambucoPrograma de Pós-Graduação em Informática AplicadaUFRPEBrasilDepartamento de Estatística e InformáticaKeystroke dynamicsModelos de seleçãoDispositivo móvelCIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAOSeleção de características aplicado ao keystroke dynamics em dispositivos móveisinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesis-8268485641417162699600600600-67745551403961205013671711205811204509info:eu-repo/semantics/openAccessreponame:Biblioteca Digital de Teses e Dissertações da UFRPEinstname:Universidade Federal Rural de Pernambuco (UFRPE)instacron:UFRPEORIGINALUrbanno Pereira de Siqueira Leite.pdfUrbanno Pereira de Siqueira Leite.pdfapplication/pdf1660810http://www.tede2.ufrpe.br:8080/tede2/bitstream/tede2/7861/2/Urbanno+Pereira+de+Siqueira+Leite.pdfe2e67d450cc32a40bfc934c706068368MD52LICENSElicense.txtlicense.txttext/plain; charset=utf-82165http://www.tede2.ufrpe.br:8080/tede2/bitstream/tede2/7861/1/license.txtbd3efa91386c1718a7f26a329fdcb468MD51tede2/78612019-02-22 12:24:50.633oai:tede2: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Biblioteca Digital de Teses e Dissertaçõeshttp://www.tede2.ufrpe.br:8080/tede/PUBhttp://www.tede2.ufrpe.br:8080/oai/requestbdtd@ufrpe.br ||bdtd@ufrpe.bropendoar:2024-05-28T12:36:12.986012Biblioteca Digital de Teses e Dissertações da UFRPE - Universidade Federal Rural de Pernambuco (UFRPE)false
dc.title.por.fl_str_mv Seleção de características aplicado ao keystroke dynamics em dispositivos móveis
title Seleção de características aplicado ao keystroke dynamics em dispositivos móveis
spellingShingle Seleção de características aplicado ao keystroke dynamics em dispositivos móveis
LEITE, Urbanno Pereira de Siqueira
Keystroke dynamics
Modelos de seleção
Dispositivo móvel
CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
title_short Seleção de características aplicado ao keystroke dynamics em dispositivos móveis
title_full Seleção de características aplicado ao keystroke dynamics em dispositivos móveis
title_fullStr Seleção de características aplicado ao keystroke dynamics em dispositivos móveis
title_full_unstemmed Seleção de características aplicado ao keystroke dynamics em dispositivos móveis
title_sort Seleção de características aplicado ao keystroke dynamics em dispositivos móveis
author LEITE, Urbanno Pereira de Siqueira
author_facet LEITE, Urbanno Pereira de Siqueira
author_role author
dc.contributor.advisor1.fl_str_mv FERREIRA, Tiago Alessandro Espínola
dc.contributor.referee1.fl_str_mv CAVALCANTI, George Darmiton da Cunha
dc.contributor.referee2.fl_str_mv GARROZI, Cícero
dc.contributor.referee3.fl_str_mv MIRANDA, Péricles Barbosa Cunha de
dc.contributor.authorLattes.fl_str_mv http://lattes.cnpq.br/3544721962219698
dc.contributor.author.fl_str_mv LEITE, Urbanno Pereira de Siqueira
contributor_str_mv FERREIRA, Tiago Alessandro Espínola
CAVALCANTI, George Darmiton da Cunha
GARROZI, Cícero
MIRANDA, Péricles Barbosa Cunha de
dc.subject.por.fl_str_mv Keystroke dynamics
Modelos de seleção
Dispositivo móvel
topic Keystroke dynamics
Modelos de seleção
Dispositivo móvel
CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
dc.subject.cnpq.fl_str_mv CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
description In this paper, feature selection models are developed for keystroke dynamics on mobile devices. Two models are elaborated, one based on Genetic Algorithm (GA) and another on PSO. The proposed selectors are applied to a public database of keystroke dynamics built from mobile devices. Both methods of selection are used in conjunction with various classifiers: Naive Bayes, Bayes Net, C4.5 (decision tree), Random Forest, k-NN, Support Vector Machine (SVM) and Multilayer Perceptron (MLP). The feature selection methods developed here are evaluated by accuracy, false positive rate (FAR), false negative rate (FRR) - all these measures are obtained from the classifications - and also by the reduction rates of characteristics. The results obtained from the execution of several experiments show that the proposed models were able to add improvements to the measures of performance - when compared to the results of the classifications without selection -, besides reaching high levels of reduction of characteristics. Through a comparative analysis it was also possible to verify that the models developed in this work have performances compatible with other selectors already available in the literature. The proposed methods also call attention for the stability of their behavior, in such a way that the results generated by them have low indices of variability. In this work it was possible to identify the most selected features and also those less chosen by the models, showing that an attribute can be quite selected for a particular classification method, but not so chosen for another classifier. Already analyzing the frequency of selection of characteristics according to their type, it was verified that the two characteristics most selected by the proposed selection methods are attributes inherent to mobile devices.
publishDate 2018
dc.date.issued.fl_str_mv 2018-02-26
dc.date.accessioned.fl_str_mv 2019-02-22T15:24:50Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
status_str publishedVersion
dc.identifier.citation.fl_str_mv LEITE, Urbanno Pereira de Siqueira. Seleção de características aplicado ao keystroke dynamics em dispositivos móveis. 2018. 109 f. Dissertação (Programa de Pós-Graduação em Informática Aplicada) - Universidade Federal Rural de Pernambuco, Recife.
dc.identifier.uri.fl_str_mv http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/7861
identifier_str_mv LEITE, Urbanno Pereira de Siqueira. Seleção de características aplicado ao keystroke dynamics em dispositivos móveis. 2018. 109 f. Dissertação (Programa de Pós-Graduação em Informática Aplicada) - Universidade Federal Rural de Pernambuco, Recife.
url http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/7861
dc.language.iso.fl_str_mv por
language por
dc.relation.program.fl_str_mv -8268485641417162699
dc.relation.confidence.fl_str_mv 600
600
600
dc.relation.department.fl_str_mv -6774555140396120501
dc.relation.cnpq.fl_str_mv 3671711205811204509
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Universidade Federal Rural de Pernambuco
dc.publisher.program.fl_str_mv Programa de Pós-Graduação em Informática Aplicada
dc.publisher.initials.fl_str_mv UFRPE
dc.publisher.country.fl_str_mv Brasil
dc.publisher.department.fl_str_mv Departamento de Estatística e Informática
publisher.none.fl_str_mv Universidade Federal Rural de Pernambuco
dc.source.none.fl_str_mv reponame:Biblioteca Digital de Teses e Dissertações da UFRPE
instname:Universidade Federal Rural de Pernambuco (UFRPE)
instacron:UFRPE
instname_str Universidade Federal Rural de Pernambuco (UFRPE)
instacron_str UFRPE
institution UFRPE
reponame_str Biblioteca Digital de Teses e Dissertações da UFRPE
collection Biblioteca Digital de Teses e Dissertações da UFRPE
bitstream.url.fl_str_mv http://www.tede2.ufrpe.br:8080/tede2/bitstream/tede2/7861/2/Urbanno+Pereira+de+Siqueira+Leite.pdf
http://www.tede2.ufrpe.br:8080/tede2/bitstream/tede2/7861/1/license.txt
bitstream.checksum.fl_str_mv e2e67d450cc32a40bfc934c706068368
bd3efa91386c1718a7f26a329fdcb468
bitstream.checksumAlgorithm.fl_str_mv MD5
MD5
repository.name.fl_str_mv Biblioteca Digital de Teses e Dissertações da UFRPE - Universidade Federal Rural de Pernambuco (UFRPE)
repository.mail.fl_str_mv bdtd@ufrpe.br ||bdtd@ufrpe.br
_version_ 1810102255852978176