{"id":588,"date":"2016-11-25T18:00:51","date_gmt":"2016-11-25T21:00:51","guid":{"rendered":"http:\/\/sbbd.org.br\/2017-teste\/?page_id=588"},"modified":"2017-10-06T08:34:35","modified_gmt":"2017-10-06T11:34:35","slug":"short-courses","status":"publish","type":"page","link":"https:\/\/sbbd.org.br\/2017\/index.php\/short-courses\/","title":{"rendered":"SHORT COURSES"},"content":{"rendered":"

[toggle title=”Short course 1: 02\/out\/2017, 08:30 \u00e0s 10:00 e 10:30 \u00e0s 12:00 – \u00c9 uma quest\u00e3o de tempo! Extraindo Conhecimento de Redes Sociais Temporais (It\u2019s a matter of time! Knowledge Discovery from Temporal Social Networks<\/em>)<\/strong>” state=”opened”]<\/p>\n

Slides<\/a><\/p>\n

Resumo: Os dados est\u00e3o estruturados na forma de rede. E agora? Como analis\u00e1-los? Extrair conhecimento desse tipo de dado n\u00e3o e\u0301 uma tarefa simples e requer o uso de ferramentas e t\u00e9cnicas adequadas, especialmente em cen\u00e1rios que levam em conta o volume de dados e o aspecto temporal da rede. Existe uma vasta literatura acerca de como coletar, pre\u0301-processar e modelar dados de mi\u0301dias sociais em forma de redes, bem como acerca das principais me\u0301tricas de centralidade. Pore\u0301m, ainda ha\u0301 muito a ser discutido em rela\u00e7\u00e3o a\u0300 ana\u0301lise da rede obtida. Neste minicurso considera-se, ent\u00e3o, que os dados ja\u0301 foram coletados e ja\u0301 est\u00e3o estruturados em forma de rede e discute-se sobre te\u0301cnicas para analisa\u0301-los, considerando especialmente a perspectiva temporal. Primeiro ser\u00e3o apresentados conceitos relacionados a\u0300 defini\u00e7\u00e3o do problema, redes temporais e me\u0301tricas para ana\u0301lise de rede. Em seguida, em um aspecto mais pra\u0301tico ser\u00e3o mostradas te\u0301cnicas de visualiza\u00e7\u00e3o e processamento de redes socias temporais. o final, tre\u0302s estudos de caso com dados de playlists de m\u00fasicas, do Twitter e de liga\u00e7\u00f5es telef\u00f4nicas ser\u00e3o discutidos, ilustrando do comec\u0327o ao fim como funciona a extra\u00e7\u00e3o de conhecimento de dados em redes sociais temporais.<\/p>\n

Abstract: Data is structured as a network. And now? How to analyze it? Extracting knowledge from network data is not a simple task and requires the use of appropriate tools and techniques, especially in scenarios that take into account the volume and evolving aspects of the network. There is a vast literature on how to collect, process and model social media data in the form of networks, as well as key metrics of centrality. However, there is still much to be discussed in relation to the analysis of the underlying network. In this short course we consider that data has already been collected and is already structured as a network. The goal is to discuss techniques to analyze these network data, especially considering the time perspective. First, concepts related to problem definition, temporal networks and metrics for network analysis will be presented. Next, in a more practical aspect will be shown techniques of visualization and processing of temporal networks. In the end, three case studies with real data from music playlists, Twitter and phone calls will be discussed, illustrating how to extract knowledge from temporal social networks.<\/p>\n

[row][column column=”one-fourth”] \"\"<\/p>\n

[\/column][column column=”three-fourth”]\u00a0Fab\u00edola S. F. Pereira<\/a>: Doutoranda em Cie\u0302ncia da Computac\u0327a\u0303o na Universidade Federal de Uberla\u0302ndia (UFU), com peri\u0301odo sandui\u0301che no LIAAD, um grupo pertencente ao INESC TEC, Portugal. Possui gradua\u00e7\u00e3o (2009) e mestrado (2011) em Ci\u00eancia da Computa\u00e7\u00e3o tamb\u00e9m pela UFU. E\u0301 autora de artigos peer-reviewed nas a\u0301reas de redes temporais, ana\u0301lise de redes sociais e prefere\u0302ncias do usua\u0301rio. Atuou como chair da special session em redes evolutivas (EvoNets) na confere\u0302ncia DSAA\u201917 [\/column][\/row]<\/p>\n

[row][column column=”one-fourth”] \"\"<\/p>\n

[\/column][column column=”three-fourth”] Jo\u00e3o Gama<\/a>: E\u0301 professor associado da Faculdade de Economia, Universidade do Porto. E\u0301 pesquisador e vice-diretor do LIAAD, um grupo pertencente ao INESC TEC. Obteve o ti\u0301tulo de Ph.D. pela Universidade do Porto em 2000. Tem trabalhado em diversos projetos nacionais e europeus em sistemas de aprendizado incremental e adaptativo, descoberta de conhecimento ubi\u0301quo, aprendizado a partir de dados massivos e em fluxo, etc. E\u0301 autor de diversos livros em Minerac\u0327a\u0303o de Dados e de mais de 250 artigos peer-reviewed nas a\u0301reas de aprendizado de ma\u0301quina, minerac\u0327a\u0303o de dados e data streams. [\/column][\/row]<\/p>\n

[row][column column=”one-fourth”] \"\"<\/p>\n

[\/column][column column=”three-fourth”] Gina M. B. de Oliveira<\/a>: Bolsista de produtividade do CNPq de 2001 a 2017 (PQ- 2). Possui graduac\u0327a\u0303o em Engenharia Ele\u0301trica pela Universidade Federal de Uberla\u0302ndia (1990), mestrado em Engenharia Eletro\u0302nica e Computac\u0327a\u0303o pelo Instituto Tecnolo\u0301gico de Aerona\u0301utica (1992) e doutorado em Engenharia Eletro\u0302nica e Computac\u0327a\u0303o pelo Instituto Tecnolo\u0301gico de Aerona\u0301utica (1999). Po\u0301s-doutorado de 07\/2013 a 07\/2014 na Heriot-Watt University (Edinburgh-Scotland) na a\u0301rea de robo\u0301tica bio-inspirada. Atualmente e\u0301 professora associada da Universidade Federal de Uberla\u0302ndia. Tem experie\u0302ncia na a\u0301rea de Cie\u0302ncia da Computac\u0327a\u0303o, atuando principalmente nos seguintes temas: algoritmos gene\u0301ticos, auto\u0302matos celulares, computac\u0327a\u0303o evolutiva, computac\u0327a\u0303o bio-inspirada, rob\u00f3tica bio-inspirada e intelige\u0302ncia artificial. [\/column][\/row]<\/p>\n

[\/toggle]<\/p>\n

[toggle title=”Short course 2: 02\/out\/2017, 08:30 \u00e0s 10:00 e 10:30 \u00e0s 12:00 – Como funciona o Deep Learning (How Deep Learning works<\/em>)<\/strong>” state=”opened”]<\/p>\n

Slides<\/a> — Material<\/a><\/p>\n

Resumo: Aprendizado profundo, como uma sub\u00e1rea de aprendizado de m\u00e1quina, utiliza da estrat\u00e9gia de criar modelos em camadas de representa\u00e7\u00f5es cujos par\u00e2metros s\u00e3o aprendidos por meio de exemplos conhecidos. A id\u00e9ia central que embasa esse tipo de t\u00e9cnica n\u00e3o \u00e9 nova, mas \u00e9 recente a fama que o cerca, causada por resultados impressionantes em particular com tarefas relacionadas \u00e0 percep\u00e7\u00e3o, essas historicamente vistas como de dif\u00edcil resolu\u00e7\u00e3o por computadores. Apesar de parecerem m\u00e9todos complexos, esses s\u00e3o na verdade compostos de elementos de processamento simples que realizam basicamente transforma\u00e7\u00f5es lineares em cadeia, mapeando subsequentes espa\u00e7os vetoriais. A partir de uma formula\u00e7\u00e3o alg\u00e9brica, esse curso apresenta como funciona o aprendizado profundo desde seus componentes b\u00e1sicos at\u00e9 os algoritmos utilizados para o aprendizado. Como casos de estudo s\u00e3o abordados os problemas de classifica\u00e7\u00e3o e aprendizado de caracter\u00edsticas em cen\u00e1rios supervisionados e n\u00e3o supervisionados utilizando redes convolucionais e auto-encoders. O objetivo \u00e9 prover entendimento do funcionamento interno desses modelos e o que os diferem de modelos n\u00e3o profundos, suas vantagens e limita\u00e7\u00f5es te\u00f3ricas, bem como instru\u00e7\u00f5es pr\u00e1ticas para aplica\u00e7\u00f5es.<\/p>\n

Abstract: Deep learning, as a subfield of machine learning, uses the strategy of creating models by stacking representation layers whose parameters are learned using known data. The central idea of this type of technique is not new, but it is recent the hype surrounding the field, caused by impressive results in particular with perception-related tasks, which were historically seen to be very difficult to be tackled by computers. Although seemingly complex methods, those are composed of simple computing elements that perform basically a chain of linear transformations, mapping subsequent vector spaces. From an algebraic formulation, this short-course presents how deep learning works from its basic components to the algorithms used to achieve learning. As case-studies the problems of classification and feature learning are presented in supervised and non-supervised scenarios using convolutional networks and auto-encoders. The objective is to provide understanding of the inner workings of such models and what makes them different from non-deep models, their theoretical advantages and limitations, as well as practical instructions for applications.<\/p>\n

[column column=”one-fourth”]<\/p>\n

\"\"<\/p>\n

[\/column][column column=”three-fourth”] Moacir Ponti<\/a> \u00e9 professor no Instituto de Ci\u00eancias Matem\u00e1ticas e de Computa\u00e7\u00e3o da Universidade de S\u00e3o Paulo (ICMC\/USP). Visitante no Centre for Vision, Speech and Signal Processing (CVSSP) da Universidade de Surrey em 2016. Possui doutorado (2008) e mestrado (2004) pela Universidade Federal de S\u00e3o Carlos. Desenvolve pesquisa nas \u00e1reas de Reconhecimento de Padr\u00f5es, Processamento de Sinais, Imagens e Video. Seus interesses de pesquisa atuais incluem Detec\u00e7\u00e3o de Anomalias em V\u00eddeos, Aprendizado de Caracter\u00edsticas Espa\u00e7o-Temporais e Busca Visual Multidom\u00ednio. [\/column]<\/p>\n

[\/toggle]<\/p>\n

[toggle title=”Short course 3: 02\/out\/2017, 13:30 \u00e0s 15:00 e 15:30 \u00e0s 17:00 – Data Analytics in Sports: Changing the game<\/strong>” state=”opened”]<\/p>\n

Slides<\/a><\/p>\n

Resumo: Nas \u00faltimas d\u00e9cadas, pesquisadores v\u00eam desenvolvendo diferentes t\u00e9cnicas para entender quais fatores influenciam os resultados esportivos e, consequentemente, qual o papel da preditibilidade e da aleatoriedade nos jogos. Com a atual evolu\u00e7\u00e3o das t\u00e9cnicas de aquisi\u00e7\u00e3o, armazenamento e processamento de grande volumes de informa\u00e7\u00f5es, as an\u00e1lises de dados ganharam ainda mais import\u00e2ncia para a descoberta de novos conhecimentos e v\u00eam transformando os comportamentos de todos envolvidos com o esporte. Este cap\u00edtulo apresenta uma introdu\u00e7\u00e3o sobre o tema atrav\u00e9s de: (i) uma discuss\u00e3o sobre a influ\u00eancia da an\u00e1lise de dados nos esportes, (ii) a apresenta\u00e7\u00e3o de estudos de casos de sucesso, (iii) uma an\u00e1lise dos processos computacionais para descoberta de conhecimento e modelagem de predi\u00e7\u00e3o, (iv) uma an\u00e1lise comparativa dos mercados de aposta e (v) uma explora\u00e7\u00e3o das oportunidades geradas nesse campo de pesquisa.<\/p>\n

Abstract: In the last decades, researchers have been developing different techniques to understand which factors influence the sporting results and, consequently, the role of predictability and randomness in the games. With the current evolution of the techniques of acquisition, storage and processing of large volumes of information, data analyzes has gained even more importance for the discovery of new knowledge and has been transforming the behaviors of all involved with the sport. This chapter presents an introduction to this topic through: (i) a discussion of the influence of data analysis in sports, (ii) the presentation of case studies of success, (iii) an analysis of computational processes for knowledge discovery and prediction modeling, (iv) a comparative analysis of the betting markets, and (v) an exploration of the opportunities generated in this field of research.<\/p>\n

[row][column column=”one-fourth”] \"\"
\n[\/column][column column=”three-fourth”] Igor Barbosa da Costa \u00e9 professor de computa\u00e7\u00e3o do Instituto Federal de Educa\u00e7\u00e3o, Ci\u00eancia e Tecnologia da Para\u00edba (IFPB), campus Campina Grande. Possui gradua\u00e7\u00e3o em Ci\u00eancia da Computa\u00e7\u00e3o pela Universidade Federal de Campina Grande – UFCG (2006) e mestrado pela Universidade Federal de Pernambuco – UFPE (2010). Atualmente \u00e9 doutorando de Ci\u00eancia da Computa\u00e7\u00e3o na UFCG e tem realizado pesquisas enolvendo minera\u00e7\u00e3o e extra\u00e7\u00e3o de conhecimento em dados de futebol. [\/column][\/row]<\/p>\n

[row][column column=”one-fourth”] \"\"
\n[\/column][column column=”three-fourth”] Carlos Eduardo Santos Pires holds a Ph.D. in Computer Science from Universidade Federal de Pernambuco (Brazil). Since 2009, he is a professor in Computer Science at the Computing and Systems Department of the Universidade Federal de Campina Grande, where he currently collaborates with research in the area of Information Systems and Databases at the Data Quality Laboratory of the Universidade Federal de Campina Grande. He has experience in computer science, with emphasis on Databases, acting on the following topics: decision support systems, knowledge discovery, data quality, and information integration. [\/column][\/row]<\/p>\n

[row][column column=”one-fourth”] \"\"
\n[\/column][column column=”three-fourth”] Leandro Marinho Balby is an assistant professor at the Department of Computer Science of the Federal University of Campina Grande (UFCG), Brazil. He holds BSc (2002) and MSc (2005) degrees in computer science and electrical engineering resp. from UFCG, and a Ph.D. (2010) in computer science from the University of Hildesheim, Germany. His research interests encompass Recommender Systems and Machine Learning (ML) in various domains, including theWeb, social media, education, economy and smart cities. As a result of his work, he has published papers in several major academic venues such as KDD, ECML\/PKDD, RecSys, Hypertext, ISMIR and ISWC. In particular, he received the best paper award at the 2015 ACM Conference on Recommender Systems. He is a regular program committee member of several premier conferences in IR and ML, including SIGIR, WWW, WSDM, CIKM, ISMIR, RecSys and SDM. In addition, he co-organized the workshops on Social Personalization (co-located with Hypertext), Social Personalization & Search (co-located with SIGIR) and served as PC chair of the 3rd and 4th Symposium on Knowledge Discovery, Mining and Learning (KDMiLe) [\/column][\/row]<\/p>\n

[\/toggle]<\/p>\n

 <\/p>\n","protected":false},"excerpt":{"rendered":"

[toggle title=”Short course 1: 02\/out\/2017, 08:30 \u00e0s 10:00 e 10:30 \u00e0s 12:00 – \u00c9 uma quest\u00e3o de tempo! Extraindo Conhecimento… <\/p>\n","protected":false},"author":6,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-588","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/sbbd.org.br\/2017\/index.php\/wp-json\/wp\/v2\/pages\/588","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sbbd.org.br\/2017\/index.php\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/sbbd.org.br\/2017\/index.php\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/sbbd.org.br\/2017\/index.php\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/sbbd.org.br\/2017\/index.php\/wp-json\/wp\/v2\/comments?post=588"}],"version-history":[{"count":49,"href":"https:\/\/sbbd.org.br\/2017\/index.php\/wp-json\/wp\/v2\/pages\/588\/revisions"}],"predecessor-version":[{"id":1581,"href":"https:\/\/sbbd.org.br\/2017\/index.php\/wp-json\/wp\/v2\/pages\/588\/revisions\/1581"}],"wp:attachment":[{"href":"https:\/\/sbbd.org.br\/2017\/index.php\/wp-json\/wp\/v2\/media?parent=588"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}