{"id":463,"date":"2021-06-27T18:12:12","date_gmt":"2021-06-27T21:12:12","guid":{"rendered":"https:\/\/www.inf.ufrgs.br\/ontobras\/?page_id=463"},"modified":"2022-01-25T15:04:20","modified_gmt":"2022-01-25T18:04:20","slug":"palestras-2021","status":"publish","type":"page","link":"https:\/\/www.inf.ufrgs.br\/ontobras\/pt\/palestras-2021\/","title":{"rendered":"Palestras 2021"},"content":{"rendered":"

Uso de Ontologias em Projetos de Data Science: Exemplos<\/em><\/h2>\n

Renato Rocha Souza<\/h3>\n

Nesta palestra, discutiremos, por meio de exemplos pr\u00e1ticos, como as ontologias podem nos ajudar no mundo da ci\u00eancia de dados.<\/p>\n

\"\"<\/div>\n

Bio:<\/strong> Renato Rocha Souza \u00e9 bacharel em Engenharia El\u00e9trica; possui mestrado em Engenharia de Produ\u00e7\u00e3o; doutorado em Ci\u00eancia da Informa\u00e7\u00e3o e tr\u00eas p\u00f3s-doutorados em Ci\u00eancia da Computa\u00e7\u00e3o. \u00c9 professor e pesquisador da Escola de Matem\u00e1tica Aplicada (EMAp) da Funda\u00e7\u00e3o Get\u00falio Vargas e professor colaborador da Escola de Ci\u00eancia da Informa\u00e7\u00e3o da Universidade Federal de Minas Gerais, al\u00e9m de fellow da Columbia University (US), University of South Wales (UK) e \u00d6sterreichische Akademie der Wissenschaften (AT). Atua na \u00e1rea de Ci\u00eancia da Informa\u00e7\u00e3o e da Computa\u00e7\u00e3o, pesquisando temas como Processamento de Linguagem Natural, Data Science e Machine Learning.<\/p>\n

 <\/p>\n

\n

\u202aThe Crisis of Content<\/em><\/h2>\n

Michael Gruninger<\/h3>\n

Although a plethora of ontologies have been developed in a wide variety of domains, there is often a sense in which it is difficult to measure progress in the field of applied ontology. In some domains there is a mindset that treats ontologies as being as arbitrary as software code, so there is no point in evaluating them, and there cannot possibly be any consensus on which ontologies to use. In other domains, there is an abundance of ontologies but no understanding of their relationships, leading to a perception of continually reinventing the wheel. Far too often, the only criteria for selecting ontologies are political, not technical. If we proceed further down this road, we ultimately risk irrelevance. Against this viewpoint, I would offer an approach to ontology design that focuses on formalizing the intended semantics of an ontology, so that sharability and reusability is guaranteed.<\/p>\n

\"\"<\/div>\n

Bio:<\/strong> Michael Gruninger is a Full Professor in the Department of Mechanical and Industrial Engineering at the University of Toronto. He has been working in the area of formal ontology and its applications to industrial problems for the past twenty years. He has published over 140 peer-reviewed papers with over 16000 citations, including a seminal paper in the methodology of ontology design and evaluation. Several of Gruninger’s projects have been adopted as international standards, including the Process Specification Language (published as International Standard ISO 18629) and Common Logic (published as International Standard ISO 24707). Gruninger is Past-President of the International Association of Ontology (IAOA) and its Applications, and is Editor-in-Chief of the journal Applied Ontology. He has been Programme Chair and Conference chair of international conferences, particularly Formal Ontology and Information Systems, which is the flagship conference for IAOA. He is also a Fellow of the Vespucci Institute for the Advancement of Geographic Information Science.<\/p>\n

\n

\u202aDesigning, Building, and Maintaining Ontology-Enabled Interdisciplinary Applications<\/em><\/h2>\n

Deborah McGuinness<\/h3>\n

There is growing interest in large interdisciplinary data and knowledge portals. With this, comes the need for ontologies that can be used by broad audiences. The ontologies typically need to leverage a range of existing ontologies, they need to evolve as those ontologies evolve, and they need to be maintained by often a diverse group of people. This talk will present some learnings from a number of large, interdisciplinary efforts and will present some methodologies and examples of ontology- and knowledge graph-enabled data portals that are collaborative designed and maintained by interdisciplinary teams.<\/p>\n

\"\"<\/div>\n

Bio: <\/strong> Deborah McGuinness is the Tetherless World Senior Constellation Chair and Professor of Computer, Cognitive, and Web Sciences at RPI. She is also the founding director of the RPI Web Science Research Center. Deborah has been recognized with awards as a fellow of the American Association for the Advancement of Science (AAAS) for contributions to the Semantic Web, knowledge representation, and reasoning environments and as the recipient of the Robert Engelmore award from the Association for the Advancement of Artificial Intelligence (AAAI) for leadership in Semantic Web research and in bridging Artificial Intelligence (AI) and eScience, significant contributions to deployed AI applications, and extensive service to the AI community. Deborah currently leads a number of large diverse data intensive resource efforts and her team is creating next generation ontology-enabled research infrastructure for work in large interdisciplinary settings. Prior to joining RPI, Deborah was the acting director of the Knowledge Systems, Artificial Intelligence Laboratory and Senior Research Scientist in the Computer Science Department of Stanford University, and previous to that she was at AT&T Bell Laboratories in Artificial Intelligence Research. Deborah also has consulted with numerous large corporations as well as emerging startup companies wishing to plan, develop, deploy, and maintain semantic web and\/or AI applications. Deborah has also worked as an expert witness in a number of cases, and has deposition and trial experience. Some areas of recent work include: data science, next generation health advisors, ontology design and evolution environments, semantically-enabled virtual observatories, semantic integration of scientific data, context-aware mobile applications, search, eCommerce, configuration, and supply chain management. Deborah holds a Bachelor of Math and Computer Science from Duke University, her Master of Computer Science from University of California at Berkeley, and her Ph.D. in Computer Science from Rutgers University.<\/p>\n

\n

Integra\u00e7\u00e3o Sem\u00e2ntica: Por que Grafo de Conhecimento Sem\u00e2ntico \u00e9 a Solu\u00e7\u00e3o?<\/em><\/h2>\n

Prof. V\u00e2nia Vidal<\/strong><\/em> e Caio Viktor<\/strong><\/em> (Universidade Federal do Cear\u00e1)<\/h3>\n

Grafo de conhecimento Sem\u00e2ntico \u00e9 um novo paradigma baseado nas tecnologias da web sem\u00e2ntica e do grafo de conhecimento (GC) para integrar fontes de dados heterog\u00eaneas. O principal objetivo de um GC Sem\u00e2ntico \u00e9 fornecer uma vis\u00e3o ontol\u00f3gica e unificada, para que aplica\u00e7\u00f5es possam ter acesso integrado aos dados das fontes atrav\u00e9s da vis\u00e3o Sem\u00e2ntica. O uso de ontologias, al\u00e9m de garantir a f\u00e1cil integra\u00e7\u00e3o de m\u00faltiplas fontes heterog\u00eaneas de dados, tamb\u00e9m prov\u00ea uma representa\u00e7\u00e3o sem\u00e2ntica formal para permitir infer\u00eancia e processamento de m\u00e1quina. Com o uso de GC sem\u00e2nticos, as empresas conseguem obter novos insights<\/em> sobre seus dados e criar aplicativos empresariais inovadores. Grafos de conhecimento sem\u00e2nticos, representados em RDF, fornecem a melhor estrutura para integra\u00e7\u00e3o sem\u00e2ntica e reutiliza\u00e7\u00e3o de dados, porque combinam: Expressividade, Desempenho, Interoperabilidade e Padroniza\u00e7\u00e3o.<\/p>\n

Nesta palestra apresentaremos os princ\u00edpios fundamentais de GC Sem\u00e2ntico, e um processo incremental para a constru\u00e7\u00e3o de um GCS. Tamb\u00e9m, abordaremos o papel de um Grafo de Conhecimento em aplica\u00e7\u00f5es de intelig\u00eancia artificial e aprendizado de m\u00e1quina. Neste contexto, ser\u00e3o apresentados desde seus usos mais tradicionais, tais como infer\u00eancias de fatos baseadas em regras declarativas e consulta sem\u00e2ntica, at\u00e9 t\u00f3picos mais recentes, tais como predi\u00e7\u00e3o de links e o uso do GC como features<\/em> para algoritmos de aprendizado de m\u00e1quina.<\/p>\n

Os Principais T\u00f3picos dessa palestra incluem:<\/p>\n