Pub Date : 2021-07-18DOI: 10.5753/SEMISH.2021.15835
Livy Real, Karina M. Johansson, Júlio C. S. Mendes, Bianca M. Lopes, Marcio T. I. Oshiro
This paper explores how Natural Language Processing techniques can be integrated to solve real-world problems in the e-commerce scenario. We address the issue of having high quality information products offered to customers in a marketplace platform, composed by thousands of sellers producing original content in multiple languages, following different SEO and cultural assumptions. We propose an NLP pipeline to generate high quality titles products in Portuguese.
{"title":"Generating E-commerce Product Titles in Portuguese","authors":"Livy Real, Karina M. Johansson, Júlio C. S. Mendes, Bianca M. Lopes, Marcio T. I. Oshiro","doi":"10.5753/SEMISH.2021.15835","DOIUrl":"https://doi.org/10.5753/SEMISH.2021.15835","url":null,"abstract":"This paper explores how Natural Language Processing techniques can be integrated to solve real-world problems in the e-commerce scenario. We address the issue of having high quality information products offered to customers in a marketplace platform, composed by thousands of sellers producing original content in multiple languages, following different SEO and cultural assumptions. We propose an NLP pipeline to generate high quality titles products in Portuguese.","PeriodicalId":206312,"journal":{"name":"Anais do XLVIII Seminário Integrado de Software e Hardware (SEMISH 2021)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123994011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-18DOI: 10.5753/SEMISH.2021.15802
Kauã B. Hopfer, Adriano Fiorese
Neste trabalho, uma forma de ranqueamento e seleção de provedores de nuvem é apresentada por meio do uso da função de Aprendizado de Máquina Automatizado (AutoML) da plataforma H2O. Ele exibe um sistema de ranqueamento que produz uma pontuação para cada provedor de nuvem avaliado, a partir da qualificação dos requisitos exigidos pelo usuário. Experimentos realizados com o auxílio da plataforma H2O, levando em consideração o treinamento e análise de modelos de regressão, apresentam resultados precisos e mais rápidos quando comparados a alternativa de resolução determinística exata.
{"title":"Uso de Aprendizado de Máquina Automatizado para Seleção de Provedores de Nuvem","authors":"Kauã B. Hopfer, Adriano Fiorese","doi":"10.5753/SEMISH.2021.15802","DOIUrl":"https://doi.org/10.5753/SEMISH.2021.15802","url":null,"abstract":"Neste trabalho, uma forma de ranqueamento e seleção de provedores de nuvem é apresentada por meio do uso da função de Aprendizado de Máquina Automatizado (AutoML) da plataforma H2O. Ele exibe um sistema de ranqueamento que produz uma pontuação para cada provedor de nuvem avaliado, a partir da qualificação dos requisitos exigidos pelo usuário. Experimentos realizados com o auxílio da plataforma H2O, levando em consideração o treinamento e análise de modelos de regressão, apresentam resultados precisos e mais rápidos quando comparados a alternativa de resolução determinística exata.","PeriodicalId":206312,"journal":{"name":"Anais do XLVIII Seminário Integrado de Software e Hardware (SEMISH 2021)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124540217","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-18DOI: 10.5753/SEMISH.2021.15829
André Ferraz, C. Ferraz
This paper argues that the essential pieces of an enduring digital identity should be privacy, security, and convenience. Authentication should be frictionless. In this sense, the core of the digital identity of the future will be created around location sensing techniques. Incognia proposes a solution to secure and frictionless authentication for mobile apps that is composed of five steps. Its proprietary technology called environment fingerprinting can identify location spoofing and precisely determine the devices actual location. Incognia has found that most mobile logins, sensitive transactions, and purchases occur at trusted locations. To date, 90% of mobile logins and 89% of mobile banking sessions happen at a trusted location. Experimental results show false-negative rates below 0.004% and a decrease of over 85% of account takeover attacks.
{"title":"Digital Identity Challenge: The Security and Convenience Dilemma","authors":"André Ferraz, C. Ferraz","doi":"10.5753/SEMISH.2021.15829","DOIUrl":"https://doi.org/10.5753/SEMISH.2021.15829","url":null,"abstract":"This paper argues that the essential pieces of an enduring digital identity should be privacy, security, and convenience. Authentication should be frictionless. In this sense, the core of the digital identity of the future will be created around location sensing techniques. Incognia proposes a solution to secure and frictionless authentication for mobile apps that is composed of five steps. Its proprietary technology called environment fingerprinting can identify location spoofing and precisely determine the devices actual location. Incognia has found that most mobile logins, sensitive transactions, and purchases occur at trusted locations. To date, 90% of mobile logins and 89% of mobile banking sessions happen at a trusted location. Experimental results show false-negative rates below 0.004% and a decrease of over 85% of account takeover attacks.","PeriodicalId":206312,"journal":{"name":"Anais do XLVIII Seminário Integrado de Software e Hardware (SEMISH 2021)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121452604","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-18DOI: 10.5753/SEMISH.2021.15807
Juliana Pereira Cabral, Herleson Paiva Pontes
Nos últimos anos, é possível observar o considerável aumento do número de ataques à segurança da informação em dispositivos móveis no Brasil, onde estima-se que ocorreram 850 mil tentativas objetivando o acesso indevido aos dados pessoais dos usuários. Este trabalho apresenta um estudo sobre a adoção dos recursos, ferramentas e boas práticas de segurança voltadas para celulares. Após a condução de um estudo de caso envolvendo 222 participantes que avaliou aspectos tecnológicos e psicológicos dos usuários acerca da segurança em seus aparelhos, os resultados sugerem que parcela considerável da população possui escasso conhecimento em relação ao emprego dos recursos e boas práticas de proteção eficientes nos dispositivos móveis, tornando-os alvos de crimes cibernéticos.
{"title":"Segurança em Dispositivos Móveis: Um Estudo Sobre a Adoção de Boas Práticas para Proteção em Celulares","authors":"Juliana Pereira Cabral, Herleson Paiva Pontes","doi":"10.5753/SEMISH.2021.15807","DOIUrl":"https://doi.org/10.5753/SEMISH.2021.15807","url":null,"abstract":"Nos últimos anos, é possível observar o considerável aumento do número de ataques à segurança da informação em dispositivos móveis no Brasil, onde estima-se que ocorreram 850 mil tentativas objetivando o acesso indevido aos dados pessoais dos usuários. Este trabalho apresenta um estudo sobre a adoção dos recursos, ferramentas e boas práticas de segurança voltadas para celulares. Após a condução de um estudo de caso envolvendo 222 participantes que avaliou aspectos tecnológicos e psicológicos dos usuários acerca da segurança em seus aparelhos, os resultados sugerem que parcela considerável da população possui escasso conhecimento em relação ao emprego dos recursos e boas práticas de proteção eficientes nos dispositivos móveis, tornando-os alvos de crimes cibernéticos.","PeriodicalId":206312,"journal":{"name":"Anais do XLVIII Seminário Integrado de Software e Hardware (SEMISH 2021)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131255637","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-18DOI: 10.5753/SEMISH.2021.15813
Nina N. Zahn, G. P. D. Molin, S. Musse
Social interactions have changed in recent years. People post their thoughts, opinions and feelings on social media platforms more often. Due to the increase in the amount of data on the internet, it is impracticable to carry out the sentiment analysis manually, requiring automation of the process. In this work, we present the corpus Cross-Media German Blog (CGB) which consists of German blogs with feelings in the domain of images, texts and posts (Ground Truth), classified according to human perceptions. We apply existing Machine Learning technologies and lexicons to the corpus to detect the feelings (negative, neutral or positive) of the images and texts and compare the results with the GT. We examined contradictory posts, when the image and text classified by humans in the same post had diverging feelings. The comparison of this article with the analysis of sentiment among the media of Brazilian blogs finds its justification for performance results in cultural differences, since, throughout this work, Brazil is classified as indulgent and Germany as a restrained country.
{"title":"Cross-Media Sentiment Analysis on German Blogs","authors":"Nina N. Zahn, G. P. D. Molin, S. Musse","doi":"10.5753/SEMISH.2021.15813","DOIUrl":"https://doi.org/10.5753/SEMISH.2021.15813","url":null,"abstract":"Social interactions have changed in recent years. People post their thoughts, opinions and feelings on social media platforms more often. Due to the increase in the amount of data on the internet, it is impracticable to carry out the sentiment analysis manually, requiring automation of the process. In this work, we present the corpus Cross-Media German Blog (CGB) which consists of German blogs with feelings in the domain of images, texts and posts (Ground Truth), classified according to human perceptions. We apply existing Machine Learning technologies and lexicons to the corpus to detect the feelings (negative, neutral or positive) of the images and texts and compare the results with the GT. We examined contradictory posts, when the image and text classified by humans in the same post had diverging feelings. The comparison of this article with the analysis of sentiment among the media of Brazilian blogs finds its justification for performance results in cultural differences, since, throughout this work, Brazil is classified as indulgent and Germany as a restrained country.","PeriodicalId":206312,"journal":{"name":"Anais do XLVIII Seminário Integrado de Software e Hardware (SEMISH 2021)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132033856","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-18DOI: 10.5753/SEMISH.2021.15815
B. Dalmoro, S. Musse
With the popularization of social networks, the sharing and consumption of content in video format becomes easier. Understanding what makes a video popular and being able to predict its popularity in number of views is useful for both content creators and advertising. In this work, we explore visual features extracted from 1,820 Facebook videos in order to predict whether they will reach more than a certain number of views on the seven days after publication. For this purpose, we used Support Vector Machine with Gaussian Radial Basis Function classification model. Using only visual features as predictors, the model with Video Characteristics and Rigidity features combined reached Kappa of 0.7324, sensitivity of 0.8930, and positive predictive value of 0.8930.
{"title":"Predicting Popularity of Facebook Videos Through Visual Features Using Support Vector Machine Classifier","authors":"B. Dalmoro, S. Musse","doi":"10.5753/SEMISH.2021.15815","DOIUrl":"https://doi.org/10.5753/SEMISH.2021.15815","url":null,"abstract":"With the popularization of social networks, the sharing and consumption of content in video format becomes easier. Understanding what makes a video popular and being able to predict its popularity in number of views is useful for both content creators and advertising. In this work, we explore visual features extracted from 1,820 Facebook videos in order to predict whether they will reach more than a certain number of views on the seven days after publication. For this purpose, we used Support Vector Machine with Gaussian Radial Basis Function classification model. Using only visual features as predictors, the model with Video Characteristics and Rigidity features combined reached Kappa of 0.7324, sensitivity of 0.8930, and positive predictive value of 0.8930.","PeriodicalId":206312,"journal":{"name":"Anais do XLVIII Seminário Integrado de Software e Hardware (SEMISH 2021)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132314312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-18DOI: 10.5753/SEMISH.2021.15820
Camila Gusmão, Karla Figueiredo, Walkir Brito
Este artigo apresenta a investigação de Técnicas de Processamento de Linguagem Natural (PLN) em Denúncias Criminais, provenientes do aplicativo do serviço do Disque Denúncia RJ para smartphone. Nele é apresentado o processo de automatização, avaliando e classificando as denúncias, objetivando reduzir o tempo de análise do conteúdo das mensagens, que possui, como principal desafio, textos escritos em linguagem muito informal, contendo muitos erros morfossintáticos. Para alcançar tais objetivos foi necessária uma investigação de técnicas de pré-processamento visando melhorar a acurácia da classificação, que foi realizada por Support Vector Machine (SVM). Os resultados encontrados são bastante promissores para o tipo de textos de denúncias, atingindo uma precisão de 76,11%.
{"title":"Técnicas de Processamento de Linguagem Natural em Denúncias Criminais: Automatização e Classificação de Texto em Português Coloquial","authors":"Camila Gusmão, Karla Figueiredo, Walkir Brito","doi":"10.5753/SEMISH.2021.15820","DOIUrl":"https://doi.org/10.5753/SEMISH.2021.15820","url":null,"abstract":"Este artigo apresenta a investigação de Técnicas de Processamento de Linguagem Natural (PLN) em Denúncias Criminais, provenientes do aplicativo do serviço do Disque Denúncia RJ para smartphone. Nele é apresentado o processo de automatização, avaliando e classificando as denúncias, objetivando reduzir o tempo de análise do conteúdo das mensagens, que possui, como principal desafio, textos escritos em linguagem muito informal, contendo muitos erros morfossintáticos. Para alcançar tais objetivos foi necessária uma investigação de técnicas de pré-processamento visando melhorar a acurácia da classificação, que foi realizada por Support Vector Machine (SVM). Os resultados encontrados são bastante promissores para o tipo de textos de denúncias, atingindo uma precisão de 76,11%.","PeriodicalId":206312,"journal":{"name":"Anais do XLVIII Seminário Integrado de Software e Hardware (SEMISH 2021)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124985463","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-18DOI: 10.5753/SEMISH.2021.15816
F. Ferreira, Bruno P. Oliveira, Rodrigo Kassick, V. Furlan, Hélio Lopes
It has been recognized that a significant increase in the production and consumption of video content occurred in the last decade. Many entertainment companies, like Globo, face challenges regarding video metadata generation. The objective of this paper is to present a suitable architecture for the Globo Group to automatically identify actors that appear in each scene of a video stream, generating new metadata annotations that can be used by recommender systems and search engines among different other applications in this industry sector.
{"title":"Scaling up Cast Face Detection in Videos at Globo","authors":"F. Ferreira, Bruno P. Oliveira, Rodrigo Kassick, V. Furlan, Hélio Lopes","doi":"10.5753/SEMISH.2021.15816","DOIUrl":"https://doi.org/10.5753/SEMISH.2021.15816","url":null,"abstract":"It has been recognized that a significant increase in the production and consumption of video content occurred in the last decade. Many entertainment companies, like Globo, face challenges regarding video metadata generation. The objective of this paper is to present a suitable architecture for the Globo Group to automatically identify actors that appear in each scene of a video stream, generating new metadata annotations that can be used by recommender systems and search engines among different other applications in this industry sector.","PeriodicalId":206312,"journal":{"name":"Anais do XLVIII Seminário Integrado de Software e Hardware (SEMISH 2021)","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122626933","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-18DOI: 10.5753/SEMISH.2021.15819
Antonio L. Amadeu, Fernando Vinturin, Guilherme A. Zimeo Morais, Maickel Hubner, E. M. Pereira, Marcelo Santos
In this work, we introduce a new methodology to discover logistic regions for pricing. We use value-based characteristics from different sources, such as demographic, socioeconomic, risk, transportation, among others, to find homogeneous and valuable pricing regions. The problem was formulated as a traditional cluster solution, where well-know metrics, such as BIC and silhouette score, were used for technical validation, and business premises and constraints, operational and sales, where used to enrich feature engineering and refine cluster formation. The results presented here are from a preliminary work that was validated through several sessions with stakeholders of interest, but it is still missing the market validation. Indeed, this work will be deployed soon and a more detailed validation process, including client adherence, will be performed and monitored until the end of this year.
{"title":"Machine Learning based Pricing Methodology for the Logistic Domain: a Preliminary Approach","authors":"Antonio L. Amadeu, Fernando Vinturin, Guilherme A. Zimeo Morais, Maickel Hubner, E. M. Pereira, Marcelo Santos","doi":"10.5753/SEMISH.2021.15819","DOIUrl":"https://doi.org/10.5753/SEMISH.2021.15819","url":null,"abstract":"In this work, we introduce a new methodology to discover logistic regions for pricing. We use value-based characteristics from different sources, such as demographic, socioeconomic, risk, transportation, among others, to find homogeneous and valuable pricing regions. The problem was formulated as a traditional cluster solution, where well-know metrics, such as BIC and silhouette score, were used for technical validation, and business premises and constraints, operational and sales, where used to enrich feature engineering and refine cluster formation. The results presented here are from a preliminary work that was validated through several sessions with stakeholders of interest, but it is still missing the market validation. Indeed, this work will be deployed soon and a more detailed validation process, including client adherence, will be performed and monitored until the end of this year.","PeriodicalId":206312,"journal":{"name":"Anais do XLVIII Seminário Integrado de Software e Hardware (SEMISH 2021)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126610212","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-18DOI: 10.5753/SEMISH.2021.15811
J. G. Jiménez, Luiz André Portes Paes Leme, M. Casanova
A knowledge base, expressed using the Resource Description Framework (RDF), can be viewed as a graph whose nodes represent entities and whose edges denote relationships. The entity relatedness problem refers to the problem of discovering and understanding how two entities are related, directly or indirectly, that is, how they are connected by paths in a knowledge base. Strategies designed to solve the entity relatedness problem typically adopt an entity similarity measure to reduce the path search space and a path ranking measure to order and filter the list of paths returned. This paper presents a framework, called CoEPinKB, that supports the empirical evaluation of such strategies. The proposed framework allows combining entity similarity and path ranking measures to generate different path search strategies. The main goals of this paper are to describe the framework and present a performance evaluation of nine different path search strategies.
{"title":"CoEPinKB: A Framework to Understand the Connectivity of Entity Pairs in Knowledge Bases","authors":"J. G. Jiménez, Luiz André Portes Paes Leme, M. Casanova","doi":"10.5753/SEMISH.2021.15811","DOIUrl":"https://doi.org/10.5753/SEMISH.2021.15811","url":null,"abstract":"A knowledge base, expressed using the Resource Description Framework (RDF), can be viewed as a graph whose nodes represent entities and whose edges denote relationships. The entity relatedness problem refers to the problem of discovering and understanding how two entities are related, directly or indirectly, that is, how they are connected by paths in a knowledge base. Strategies designed to solve the entity relatedness problem typically adopt an entity similarity measure to reduce the path search space and a path ranking measure to order and filter the list of paths returned. This paper presents a framework, called CoEPinKB, that supports the empirical evaluation of such strategies. The proposed framework allows combining entity similarity and path ranking measures to generate different path search strategies. The main goals of this paper are to describe the framework and present a performance evaluation of nine different path search strategies.","PeriodicalId":206312,"journal":{"name":"Anais do XLVIII Seminário Integrado de Software e Hardware (SEMISH 2021)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134224932","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：