New histogram-based user and item profiles for recommendation systems
Autor(a) principal: | |
---|---|
Data de Publicação: | 2021 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UFPE |
Texto Completo: | https://repositorio.ufpe.br/handle/123456789/44841 |
Resumo: | Recommendation systems play an important role in businesses such as e-commerce, digital entertainment and online education. Most recommendation systems are implemented using numerical or categorical data, that is, traditional data. This type of data can be a limiting factor when used to model complex concepts where there is internal variability or internal structure in the data. In order to overcome these limitations, symbolic data are used, where values can be intervals, probability distributions or lists of values. Symbolic data can benefit recommendation systems and this work introduces a methodology to construct recommendation systems using symbolic descriptions for users and items. The proposed methodology can be applied in the implementation of recommendation systems based on content or based on collaborative filtering. In the content-based approach, user profiles and item profiles are created from symbolic descriptions of their features and a list of items are matched against a user profile. In the approach based on collaborative filtering, user profiles are built and users are grouped to form a neighborhood, products rated by users of this neighborhood are recommended based on the similarity between the neighbor and the user who will receive the recommendation. Experiments are carried out to evaluate the effectiveness of the methodology proposed in this work in relation to existing methodologies in the literature for the two recommendation system approaches. In the experiments, it was shown that the methodology proposed in this work is able to produce ranked lists with higher quality than the methodologies in the literature, i.e., lists where items with greater relevance appear in the first positions. A movie domain dataset is used in these experiments and their results show the usefulness of the proposed methodology. |
id |
UFPE_7a9aed9537a74e904cdba82fa816d220 |
---|---|
oai_identifier_str |
oai:repositorio.ufpe.br:123456789/44841 |
network_acronym_str |
UFPE |
network_name_str |
Repositório Institucional da UFPE |
repository_id_str |
2221 |
spelling |
SAMPAIO NETO, Delmiro Daladierhttp://lattes.cnpq.br/7229583933046960http://lattes.cnpq.br/9289080285504453http://lattes.cnpq.br/4640945954423515SOUZA, Renata Maria Cardoso Rodrigues deSILVA FILHO, Telmo de Menezes e2022-06-22T18:48:53Z2022-06-22T18:48:53Z2021-12-07SAMPAIO NETO, Delmiro Daladier. New histogram-based user and item profiles for recommendation systems. 2021. Dissertação (Mestrado em Ciência da Computação) – Universidade Federal de Pernambuco, Recife, 2021.https://repositorio.ufpe.br/handle/123456789/44841Recommendation systems play an important role in businesses such as e-commerce, digital entertainment and online education. Most recommendation systems are implemented using numerical or categorical data, that is, traditional data. This type of data can be a limiting factor when used to model complex concepts where there is internal variability or internal structure in the data. In order to overcome these limitations, symbolic data are used, where values can be intervals, probability distributions or lists of values. Symbolic data can benefit recommendation systems and this work introduces a methodology to construct recommendation systems using symbolic descriptions for users and items. The proposed methodology can be applied in the implementation of recommendation systems based on content or based on collaborative filtering. In the content-based approach, user profiles and item profiles are created from symbolic descriptions of their features and a list of items are matched against a user profile. In the approach based on collaborative filtering, user profiles are built and users are grouped to form a neighborhood, products rated by users of this neighborhood are recommended based on the similarity between the neighbor and the user who will receive the recommendation. Experiments are carried out to evaluate the effectiveness of the methodology proposed in this work in relation to existing methodologies in the literature for the two recommendation system approaches. In the experiments, it was shown that the methodology proposed in this work is able to produce ranked lists with higher quality than the methodologies in the literature, i.e., lists where items with greater relevance appear in the first positions. A movie domain dataset is used in these experiments and their results show the usefulness of the proposed methodology.Os sistemas de recomendação desempenham um papel importante em negócios como e- commerce, entretenimento digital e educação online. A maioria dos sistemas de recomendação são implementados usando dados numéricos ou categóricos, ou seja, dados tradicionais. Esse tipo de dado pode ser um fator limitante quando usado para modelar conceitos complexos onde há variabilidade interna ou estrutura interna nos dados. Para superar essas limitações, são utilizados dados simbólicos, onde os valores podem ser intervalos, distribuições de prob- abilidade ou listas de valores. Dados simbólicos podem beneficiar sistemas de recomendação e este trabalho apresenta uma metodologia para construir sistemas de recomendação usando descrições simbólicas para usuários e itens. A metodologia proposta pode ser aplicada na implementação de sistemas de recomen- dação baseados em conteúdo ou baseados em filtragem colaborativa. Na abordagem baseada em conteúdo, perfis de usuários e perfis de itens são criados a partir de descrições simbólicas de seus recursos e uma lista de itens é comparada a um perfil de usuário. Na abordagem baseada em filtragem colaborativa, os perfis dos usuários são construídos e os usuários são agrupados para formar uma vizinhança, os produtos avaliados pelos usuários desta vizinhança são recomendados com base na semelhança entre o vizinho e o usuário que receberá a re- comendação. Experimentos são realizados para avaliar a eficácia da metodologia proposta neste trabalho em relação às metodologias existentes na literatura para as duas abordagens de sistema de recomendação. Nos experimentos, foi mostrado que a metodologia proposta neste trabalho é capaz de produzir listas ordenadas com qualidade superior às metodologias da lit- eratura, ou seja, listas onde os itens com maior relevância aparecem nas primeiras posições. Um conjunto de dados de domínio de filme é usado nesses experimentos e seus resultados mostram a utilidade da metodologia proposta.engUniversidade Federal de PernambucoPrograma de Pos Graduacao em Ciencia da ComputacaoUFPEBrasilhttp://creativecommons.org/licenses/by-nc-nd/3.0/br/info:eu-repo/semantics/openAccessInteligência computacionalSistemas de recomendaçãoDados simbólicosHistogramasNew histogram-based user and item profiles for recommendation systemsinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesismestradoreponame:Repositório Institucional da UFPEinstname:Universidade Federal de Pernambuco (UFPE)instacron:UFPETEXTDISSERTAÇÃO Delmiro Daladier Sampaio Neto.pdf.txtDISSERTAÇÃO Delmiro Daladier Sampaio Neto.pdf.txtExtracted texttext/plain127525https://repositorio.ufpe.br/bitstream/123456789/44841/4/DISSERTA%c3%87%c3%83O%20Delmiro%20Daladier%20Sampaio%20Neto.pdf.txt20efc0ac005120960b016bd452e19e08MD54THUMBNAILDISSERTAÇÃO Delmiro Daladier Sampaio Neto.pdf.jpgDISSERTAÇÃO Delmiro Daladier Sampaio Neto.pdf.jpgGenerated Thumbnailimage/jpeg1231https://repositorio.ufpe.br/bitstream/123456789/44841/5/DISSERTA%c3%87%c3%83O%20Delmiro%20Daladier%20Sampaio%20Neto.pdf.jpg1fa71ce4fecbd635cfc0b39e34ee3538MD55CC-LICENSElicense_rdflicense_rdfapplication/rdf+xml; charset=utf-8811https://repositorio.ufpe.br/bitstream/123456789/44841/2/license_rdfe39d27027a6cc9cb039ad269a5db8e34MD52LICENSElicense.txtlicense.txttext/plain; charset=utf-82142https://repositorio.ufpe.br/bitstream/123456789/44841/3/license.txt6928b9260b07fb2755249a5ca9903395MD53ORIGINALDISSERTAÇÃO Delmiro Daladier Sampaio Neto.pdfDISSERTAÇÃO Delmiro Daladier Sampaio Neto.pdfapplication/pdf979103https://repositorio.ufpe.br/bitstream/123456789/44841/1/DISSERTA%c3%87%c3%83O%20Delmiro%20Daladier%20Sampaio%20Neto.pdfbf291bdd119059424336ace228f5bd02MD51123456789/448412022-06-23 02:22:16.844oai:repositorio.ufpe.br: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ório InstitucionalPUBhttps://repositorio.ufpe.br/oai/requestattena@ufpe.bropendoar:22212022-06-23T05:22:16Repositório Institucional da UFPE - Universidade Federal de Pernambuco (UFPE)false |
dc.title.pt_BR.fl_str_mv |
New histogram-based user and item profiles for recommendation systems |
title |
New histogram-based user and item profiles for recommendation systems |
spellingShingle |
New histogram-based user and item profiles for recommendation systems SAMPAIO NETO, Delmiro Daladier Inteligência computacional Sistemas de recomendação Dados simbólicos Histogramas |
title_short |
New histogram-based user and item profiles for recommendation systems |
title_full |
New histogram-based user and item profiles for recommendation systems |
title_fullStr |
New histogram-based user and item profiles for recommendation systems |
title_full_unstemmed |
New histogram-based user and item profiles for recommendation systems |
title_sort |
New histogram-based user and item profiles for recommendation systems |
author |
SAMPAIO NETO, Delmiro Daladier |
author_facet |
SAMPAIO NETO, Delmiro Daladier |
author_role |
author |
dc.contributor.authorLattes.pt_BR.fl_str_mv |
http://lattes.cnpq.br/7229583933046960 |
dc.contributor.advisorLattes.pt_BR.fl_str_mv |
http://lattes.cnpq.br/9289080285504453 |
dc.contributor.advisor-coLattes.pt_BR.fl_str_mv |
http://lattes.cnpq.br/4640945954423515 |
dc.contributor.author.fl_str_mv |
SAMPAIO NETO, Delmiro Daladier |
dc.contributor.advisor1.fl_str_mv |
SOUZA, Renata Maria Cardoso Rodrigues de |
dc.contributor.advisor-co1.fl_str_mv |
SILVA FILHO, Telmo de Menezes e |
contributor_str_mv |
SOUZA, Renata Maria Cardoso Rodrigues de SILVA FILHO, Telmo de Menezes e |
dc.subject.por.fl_str_mv |
Inteligência computacional Sistemas de recomendação Dados simbólicos Histogramas |
topic |
Inteligência computacional Sistemas de recomendação Dados simbólicos Histogramas |
description |
Recommendation systems play an important role in businesses such as e-commerce, digital entertainment and online education. Most recommendation systems are implemented using numerical or categorical data, that is, traditional data. This type of data can be a limiting factor when used to model complex concepts where there is internal variability or internal structure in the data. In order to overcome these limitations, symbolic data are used, where values can be intervals, probability distributions or lists of values. Symbolic data can benefit recommendation systems and this work introduces a methodology to construct recommendation systems using symbolic descriptions for users and items. The proposed methodology can be applied in the implementation of recommendation systems based on content or based on collaborative filtering. In the content-based approach, user profiles and item profiles are created from symbolic descriptions of their features and a list of items are matched against a user profile. In the approach based on collaborative filtering, user profiles are built and users are grouped to form a neighborhood, products rated by users of this neighborhood are recommended based on the similarity between the neighbor and the user who will receive the recommendation. Experiments are carried out to evaluate the effectiveness of the methodology proposed in this work in relation to existing methodologies in the literature for the two recommendation system approaches. In the experiments, it was shown that the methodology proposed in this work is able to produce ranked lists with higher quality than the methodologies in the literature, i.e., lists where items with greater relevance appear in the first positions. A movie domain dataset is used in these experiments and their results show the usefulness of the proposed methodology. |
publishDate |
2021 |
dc.date.issued.fl_str_mv |
2021-12-07 |
dc.date.accessioned.fl_str_mv |
2022-06-22T18:48:53Z |
dc.date.available.fl_str_mv |
2022-06-22T18:48:53Z |
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 |
SAMPAIO NETO, Delmiro Daladier. New histogram-based user and item profiles for recommendation systems. 2021. Dissertação (Mestrado em Ciência da Computação) – Universidade Federal de Pernambuco, Recife, 2021. |
dc.identifier.uri.fl_str_mv |
https://repositorio.ufpe.br/handle/123456789/44841 |
identifier_str_mv |
SAMPAIO NETO, Delmiro Daladier. New histogram-based user and item profiles for recommendation systems. 2021. Dissertação (Mestrado em Ciência da Computação) – Universidade Federal de Pernambuco, Recife, 2021. |
url |
https://repositorio.ufpe.br/handle/123456789/44841 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
http://creativecommons.org/licenses/by-nc-nd/3.0/br/ info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
http://creativecommons.org/licenses/by-nc-nd/3.0/br/ |
eu_rights_str_mv |
openAccess |
dc.publisher.none.fl_str_mv |
Universidade Federal de Pernambuco |
dc.publisher.program.fl_str_mv |
Programa de Pos Graduacao em Ciencia da Computacao |
dc.publisher.initials.fl_str_mv |
UFPE |
dc.publisher.country.fl_str_mv |
Brasil |
publisher.none.fl_str_mv |
Universidade Federal de Pernambuco |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional da UFPE instname:Universidade Federal de Pernambuco (UFPE) instacron:UFPE |
instname_str |
Universidade Federal de Pernambuco (UFPE) |
instacron_str |
UFPE |
institution |
UFPE |
reponame_str |
Repositório Institucional da UFPE |
collection |
Repositório Institucional da UFPE |
bitstream.url.fl_str_mv |
https://repositorio.ufpe.br/bitstream/123456789/44841/4/DISSERTA%c3%87%c3%83O%20Delmiro%20Daladier%20Sampaio%20Neto.pdf.txt https://repositorio.ufpe.br/bitstream/123456789/44841/5/DISSERTA%c3%87%c3%83O%20Delmiro%20Daladier%20Sampaio%20Neto.pdf.jpg https://repositorio.ufpe.br/bitstream/123456789/44841/2/license_rdf https://repositorio.ufpe.br/bitstream/123456789/44841/3/license.txt https://repositorio.ufpe.br/bitstream/123456789/44841/1/DISSERTA%c3%87%c3%83O%20Delmiro%20Daladier%20Sampaio%20Neto.pdf |
bitstream.checksum.fl_str_mv |
20efc0ac005120960b016bd452e19e08 1fa71ce4fecbd635cfc0b39e34ee3538 e39d27027a6cc9cb039ad269a5db8e34 6928b9260b07fb2755249a5ca9903395 bf291bdd119059424336ace228f5bd02 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 MD5 MD5 |
repository.name.fl_str_mv |
Repositório Institucional da UFPE - Universidade Federal de Pernambuco (UFPE) |
repository.mail.fl_str_mv |
attena@ufpe.br |
_version_ |
1802310751482806272 |