Palestras 2021

Uso de Ontologias em Projetos de Data Science: Exemplos

Renato Rocha Souza

Nesta palestra, discutiremos, por meio de exemplos práticos, como as ontologias podem nos ajudar no mundo da ciência de dados.

Bio: Renato Rocha Souza é bacharel em Engenharia Elétrica; possui mestrado em Engenharia de Produção; doutorado em Ciência da Informação e três pós-doutorados em Ciência da Computação. É professor e pesquisador da Escola de Matemática Aplicada (EMAp) da Fundação Getúlio Vargas e professor colaborador da Escola de Ciência da Informação da Universidade Federal de Minas Gerais, além de fellow da Columbia University (US), University of South Wales (UK) e Österreichische Akademie der Wissenschaften (AT). Atua na área de Ciência da Informação e da Computação, pesquisando temas como Processamento de Linguagem Natural, Data Science e Machine Learning.

 


‪The Crisis of Content

Michael Gruninger

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.

Bio: 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.


‪Designing, Building, and Maintaining Ontology-Enabled Interdisciplinary Applications

Deborah McGuinness

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.

Bio: 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.


Integração Semântica: Por que Grafo de Conhecimento Semântico é a Solução?

Prof. Vânia Vidal e Caio Viktor (Universidade Federal do Ceará)

Grafo de conhecimento Semântico é um novo paradigma baseado nas tecnologias da web semântica e do grafo de conhecimento (GC) para integrar fontes de dados heterogêneas. O principal objetivo de um GC Semântico é fornecer uma visão ontológica e unificada, para que aplicações possam ter acesso integrado aos dados das fontes através da visão Semântica. O uso de ontologias, além de garantir a fácil integração de múltiplas fontes heterogêneas de dados, também provê uma representação semântica formal para permitir inferência e processamento de máquina. Com o uso de GC semânticos, as empresas conseguem obter novos insights sobre seus dados e criar aplicativos empresariais inovadores. Grafos de conhecimento semânticos, representados em RDF, fornecem a melhor estrutura para integração semântica e reutilização de dados, porque combinam: Expressividade, Desempenho, Interoperabilidade e Padronização.

Nesta palestra apresentaremos os princípios fundamentais de GC Semântico, e um processo incremental para a construção de um GCS. Também, abordaremos o papel de um Grafo de Conhecimento em aplicações de inteligência artificial e aprendizado de máquina. Neste contexto, serão apresentados desde seus usos mais tradicionais, tais como inferências de fatos baseadas em regras declarativas e consulta semântica, até tópicos mais recentes, tais como predição de links e o uso do GC como features para algoritmos de aprendizado de máquina.

Os Principais Tópicos dessa palestra incluem:

  • Integração Semântica & Grafo de Conhecimento Semântico
  • Grafo de conhecimento Semântico: Enfoque Virtual ou Materializado?
  • Plataforma de GC Semântico baseado no enfoque híbrido
  • Enfoque Incremental baseado nas tecnologias da Web Semântica para a construção de um GC Semântico
  • O Grafo de Conhecimento Semântico da SEFAZ-MA
  • Acesso integrado aos Dados do GC Semântico através da ontologia;
  • O que é Grafo de Conhecimento de Metadados, e quais seus benefícios na Gestão dos Dados.
  • Vocabulário para representação dos Metadados da Visão Semântica
  • O Papel de Grafo de Conhecimento Semântico em AI/M