Antonio Rafael Sabino Parmezan is a Ph.D. candidate of Computer Science and Computational Mathematics at the Institute of Mathematics and Computer Science from the University of São Paulo - USP - São Carlos, Brazil. He holds an M.Sc. degree in Computer Science and Computational Mathematics in USP (2016), and a B.Sc. in Computer Science from Western Paraná State University - UNIOESTE - Foz do Iguaçu, Brazil (2012). From December 2018 to May 2019, Antonio was a visiting Ph.D. student at the University of Helsinki - UH - Helsinki, Finland. He is currently a collaborator at two research centers: Laboratory of Computational Intelligence - LABIC - at USP, and Laboratory of Bioinformatics - LABI - at UNIOESTE. His research interests include data mining, machine learning, time series analysis, data stream processing, feature selection, and metalearning.

Education
2016 Ph.D. candidate in Computer Science and Computational Mathematics
University of São Paulo - USP - São Carlos, Brazil
with a research internship at the University of Helsinki - UH - Helsinki, Finland (Co-advisor: Indrė Žliobaitė)
Advisor: Gustavo Enrique de Almeida Prado Alves Batista
2013 - 2016 M.Sc. in Computer Science and Computational Mathematics
University of São Paulo - USP - São Carlos, Brazil
Advisor: Gustavo Enrique de Almeida Prado Alves Batista
2009 - 2012 B.Sc. in Computer Science
Western Paraná State University - UNIOESTE - Foz do Iguaçu, Brazil
Advisor: Huei Diana Lee
Research Projects
2019 - Present Investigation of Methods for Feature Extraction and Selection for Medical Image Analysis

O diagnóstico precoce constitui uma ferramenta importante no prognóstico e no tratamento de câncer, o qual tem sido beneficiado por meio do trabalho conjunto entre as áreas computacional e da saúde. Como exemplos de técnicas utilizadas, destacam-se os métodos de processamento de imagens digitais e de aprendizado de máquina. A combinação desses métodos permite a extração de características relevantes, as quais podem contribuir para a geração de modelos computacionais (caracterização) capazes de classificar a existência ou não de câncer em uma determinada imagem. Um dos motivos do favorecimento dessas técnicas, principalmente as de aprendizado de máquina, é o aumento e a disponibilização de bases de dados geradas na área médica. No entanto, um aspecto importante está relacionado às características usadas para representar essas imagens, as quais são fundamentais para o sucesso na construção de modelos representativos. Portanto, neste projeto serão investigados e aplicados métodos com o objetivo de auxiliar a caracterização de imagens médicas, com e sem lesões, por meio da extração de características, seleção e combinação de atributos importantes, bem como a avaliação desses conjuntos de atributos por meio da construção de modelos computacionais.

Participants: Antonio Rafael Sabino Parmezan / Huei Diana Lee - Principal Investigator / Feng Chung Wu / Newton Spolaôr / Ana Isabel Gonçalves Mendes / Rui Manuel Fonseca-Pinto / Conceição Veloso Nogueira / Matheus Maciel.
Funding Agency: Araucaria's Foundation for Supporting Scientific and Technological Development of Paraná - FAADCT/PR.
2018 - Present Hierarchical Classification of Nonstationary Data Streams Using Top-down and Big-bang Approaches
Project's Website

Most of the modern measurement devices produce data continuously in a vast volume and high arrival rate. This dynamic scenario, with a nonstationary characteristic, has been explored by the recent and expanding area of Data Stream Mining. Among the tasks supported by this field, classification is one of the most thought-provoking. This assertion is supported by the fact that several problems related to real-time data labeling have open questions, such as those involving concept drift, label latency, and open-set classification. When the labels of the events in a Data Stream are naturally arranged in a hierarchy, the classification task is even more challenging. The high effort required to include the class hierarchy into the process of labeling Data Streams is compensated by the flexibility and robustness provided to the classification process. In this project, we will investigate the use of top-down and big-bang approaches in the application of symbolic Machine Learning methods to be developed for the Hierarchical Classification of Nonstationary Data Streams. In addition to expanding state-of-the-art by proposing a novel category of hierarchical classification algorithms for the Data Streams context, this research also aims to contribute to an application of social and environmental impact for automatic insect identification.

Participants: Antonio Rafael Sabino Parmezan / Gustavo Enrique de Almeida Prado Alves Batista - Principal Investigator / Indrė Žliobaitė.
Funding Agency: Coordination for the Improvement of Higher Education Personnel - CAPES; Ph.D. Sandwich Program Abroad - PDSE.
2017 - 2019 Intelligent Traps and Sensors: an Innovative Approach to Control Insect Pests and Disease Vectors

Insects are undoubtedly important to agriculture, the environment and human health. Many insect species are beneficial to the environment and humans. For example, insects are responsible for pollinating at least two-thirds of all food consumed in the world. Due to its importance to humans, the recent decline in populations of pollinator insects, especially bees, is considered a serious environmental problem; frequently associated with pesticide exposure. In contrast, insect pests destroy over 40 billion U.S. dollars worth of food each year and vectors are responsible for spreading diseases that kill over one million people annually, such as malaria, dengue and chikungunya fevers and zika virus. In this project, we propose an intelligent trap that captures harmful insect species. Such a trap uses a sensor that we have developed over the last years to automatically recognize insect species using wingbeat data. The insect recognition will allow the creation of real-time insect density maps that can be used to support local interventions. For instance, in the case of insect pests, these maps will allow more local use of insecticides and, therefore, a reduced impact over the environment. In the case of disease vectors, this trap will make some sophisticated but highly costly interventions, such as SIT (Sterile Insect Technique), more cost-effective. In this project, we show how this real application can expand the limits of the state-of-the-art research in Computer Science, particularly in Machine Learning and Data Stream Mining areas. In order to demonstrate the practical aspects of our proposal, we will concentrate in the identification of two species: the Asian citrus psyllid, vector of greening, a terrible citrus disease and the Aedes aegypti vector of dengue, chikungunya and yellow fevers, as well as, the zika virus, recently associated with cases of microcephaly in newborns.

Participants: Antonio Rafael Sabino Parmezan / André Gustavo Maletzke / Gustavo Enrique de Almeida Prado Alves Batista - Principal Investigator / Diego Furtado Silva / Agenor Mafra Neto / Claudia Regina Milaré / Eamonn John Keogh / Juliano José Corbi / Pedro Takao Yamamoto / Ronaldo Cristiano Prati / Vinícius Mourão Alves de Souza.
Funding Agency: São Paulo Research Foundation - FAPESP.
2016 - Present Machine Learning in Dermoscopy

This international collaboration project aims to optimize usual machine learning algorithms from different paradigms to establish its optimal performance for the daily use in Dermatology. The results of this research will approach clinical users to the newly developed methods in machine learning to access diagnosis.

Participants: Antonio Rafael Sabino Parmezan / Huei Diana Lee - Principal Investigator / Feng Chung Wu / Newton Spolaôr / Narco Afonso Ravazzoli Maciejewski / Ana Isabel Gonçalves Mendes / Rui Manuel Fonseca-Pinto.
2016 - Present Hierarchical Classification of Data Streams
Project Poster

In Data Stream Mining, state-of-the-art Machine Learning algorithms for the classification task are characterized by associating each event with a class belonging to a finite, devoid of structural dependencies and usually small, set of classes. However, there are more complex dynamic problems where the classes to be predicted are naturally arranged in a hierarchy. An advantage in assigning examples to hierarchically organized classes is that the closer to the root of the hierarchy a linkage occurs, the smaller the classifier error rate tends to be. The freedom to perform a more generic classification, but with greater reliability, gives the process a versatility that is desired in many real world applications. In this project, the aim is to investigate incremental Machine Learning methods for the task of Hierarchical Classification of Nonstationary Data Streams and with restrictions concerning the availability of labels. Besides seeking to advance the topics of Hierarchical Classification and Data Stream Mining, this research proposal also intends to collaborate with a relevant and promising application for automatic insect identification.

Participants: Antonio Rafael Sabino Parmezan / Gustavo Enrique de Almeida Prado Alves Batista - Principal Investigator.
Funding Agency: Brazilian National Counsel of Technological and Scientific Development - CNPq.
2014 - 2016 Machine Learning for WebSensors: Algorithms and Applications

The popularization of textual content published in web platforms has motivated the development of methods for automatic knowledge extraction from texts. Particularly, a new range of applications and studies have been proposed to use the web as a powerful "social sensor". This allows to identify and monitor events published in news portals and social networks as epidemics detection, sentiment analysis, and political and economic indicators. Currently, the construction of a websensor is a complex task, since it depends of domain specialists to define the parameters of the sensors, i.e., search queries, filters and monitoring textual content from web. Moreover, for some problems there is no comprehension about the phenomenons to monitor, which limits the application of websensors. In this research project we investigate the use of machine learning methods to support the building of websensors. The basic idea is to use a sample of textual document from a problem and apply semi/non supervised learning methods to extract patterns from texts and support the generation of websensors. Thus, we hope to reduce the dependency of specialist domains to define parameters for websensor. Besides, each websensor represents a phenomenon related to a problem and it can be monitored during the time to be used as support to decision making.

Participants: Antonio Rafael Sabino Parmezan / Gustavo Enrique de Almeida Prado Alves Batista / Bruno Magalhães Nogueira / Camila Vaccari Sundermann / Diego Furtado Silva / Fabiano Fernandes dos Santos / Ivone Penque Matsuno / Rafael Geraldeli Rossi / Renan de Padua / Ricardo Marcondes Marcacini / Roberta Akemi Sinoara / Solange Oliveira Rezende - Principal Investigator / Tatiana Ximenes / Veronica Oliveira de Carvalho.
Funding Agency: São Paulo Research Foundation - FAPESP.
2013 - 2015 Similarity-based Time Series Prediction

In the last decade we have seen a huge increase of interest for time series methods in Data Mining and Knowledge Discovery. Such interest has culminated in the proposal of literally hundreds of methods for tasks such as classification, clustering, anomaly detection, motif detection, among others. Empirical research has demonstrated that time series similarity-based methods provide very competitive results, frequently outperforming more complex methods for tasks such as classification, clustering, and anomaly detection. We believe that the superiority of similarity-based methods is largely due to the community incessant work on distance invariances such as warping, baseline, occlusion and rotation. In contrast, statistical methods based on autoregression and moving averages are considered the state-of-the-art for time series modeling and prediction for over half-century now. In this research proposal, we ask if it really is the case, or if the Data Mining community has more to offer. In particular, we raise the hypothesis that similarity-based methods are simple, effective and parameter-light approaches for time series prediction. Although, similarity-based time series prediction methods have been researched in the recent past, we believe that previous research has failed to identify the correct invariances required for this task. The central hypothesis of this work is that only the right combination of offset, amplitude and recently-proposed complexity invariance, combined with a policy to avoid trivial matches leads to precise and meaningful predictions.

Participants: Antonio Rafael Sabino Parmezan / Gustavo Enrique de Almeida Prado Alves Batista - Principal Investigator.
Funding Agency: São Paulo Research Foundation - FAPESP.
2011 - 2012 Metalearning for Choosing Feature Selection Algorithms in Data Mining

Neste projeto, métodos para Meta-aprendizado empregados, usualmente, para relacionar o comportamento de algoritmos de indução de modelos (extração de padrões) com as diversas propriedades dos conjuntos de dados, serão estudados e utilizados para o problema de Seleção de Atributos em continuidade a trabalhos anteriores, visando à recomendação efetiva de algoritmos por meio da determinação de meta-modelos.

Participants: Antonio Rafael Sabino Parmezan / Huei Diana Lee - Principal Investigator / Feng Chung Wu.
Funding Agency: Universidade Estadual do Oeste do Paraná - UNIOESTE; Institutional Scientific Initiation Scholarship Program - PIBIC.
2010 - 2013 Colonoscopy Image Analysis using Texture Features and Methods of Computational Intelligence

Neste projeto serão pesquisados métodos para a análise de imagens médicas, no intuito de analisar as imagens provenientes de exames de coloscopia e extrair atributos que permitam a representação e determinação das anormalidades contidas nessas imagens. A partir disso, serão utilizados métodos da área de Inteligência Computacional para a construção de modelos que permitam representar padrões pictóricos. Os métodos utilizados durante o desenvolvimento do projeto serão também implementados dentro de um sistema computacional que foi desenvolvido como resultado de projeto anterior apoiado pela Fundação Araucária.

Participants: Antonio Rafael Sabino Parmezan / Carlos Andrés Ferrero / Huei Diana Lee - Principal Investigator / Willian Zalewski / André Gustavo Maletzke / Feng Chung Wu / Cláudio Saddy Rodrigues Coy / Renato Bobsin Machado / João José Fagundes / Jefferson Tales Oliva / Luiz Henrique Dutra da Costa.
Funding Agency: Araucaria's Foundation for Supporting Scientific and Technological Development of Paraná - FAADCT/PR.
2010 - 2011 Study of Importance Measures and Feature Selection Algorithms for Data Mining

Neste projeto serão pesquisadas medidas de avaliação de importância de atributos e algoritmos para a tarefa de Seleção de Atributos. Para tanto, serão estudadas e comparadas diferentes tipos de medidas de avaliação de importância de atributos pertencentes às categorias: clássica (informação, distância e dependência), consistência e precisão. Essas medidas e os algoritmos estudados, com foco na abordagem filtro, serão aplicados, principalmente, a bases de dados naturais.

Participants: Antonio Rafael Sabino Parmezan / Huei Diana Lee - Principal Investigator / Feng Chung Wu.
Funding Agency: Brazilian National Counsel of Technological and Scientific Development - CNPq; Institutional Scientific Initiation Scholarship Program - PIBIC.
2010 - Present Intelligent Data Analysis

Este projeto tem como objetivo desenvolver metodologias e ferramentas que apoiem a análise inteligente de dados com foco na área biomédica. Três dos principais temas incluem: (1) mapeamento de informações, a partir de laudos médicos e formulários contendo informações médicas, para bases de dados estruturadas; (2) extração de características e análise de imagens biomédicas; e (3) modelagem e análise de dados biomecânicos.

Participants: Antonio Rafael Sabino Parmezan / Huei Diana Lee - Principal Investigator / Feng Chung Wu / Newton Spolaôr / Gustavo Enrique de Almeida Prado Alves Batista / Cláudio Saddy Rodrigues Coy / João José Fagundes / Maria de Lourdes Setsuko Ayrizono / Jefferson Tales Oliva / Raquel Franco Leal / Solange Oliveira Rezende / Maria Carolina Monard / Leonilda Correa dos Santos / Weber Shoity Resende Takaki / Moacir Fonteque Júnior / Silvani Weber da Silva Borges / Leandro Augusto Ensina / Alice Mioranza de Almeida / Fabiano Silva / Paulo Cesar Marques Filho / Igor Utzig Picco / Thiago Ferreira de Toledo.
Publications
Journal Publications
1.
  1. Antonio Rafael Sabino Parmezan, Vinícius Mourão Alves Souza and Gustavo Enrique Almeida Prado Alves Batista. Evaluation of statistical and machine learning models for time series prediction: Identifying the state-of-the-art and the best conditions for the use of each model. Information Sciences 484:302–337, May 2019. URL, DOI, Supplementary Material BibTeX

    @article{Art:Parmezan:INS:2019:Evaluation,
    	author = "Parmezan, Antonio Rafael Sabino and Souza, Vinícius Mourão Alves and Batista, Gustavo Enrique Almeida Prado Alves",
    	title = "Evaluation of statistical and machine learning models for time series prediction: Identifying the state-of-the-art and the best conditions for the use of each model",
    	journal = "Information Sciences",
    	publisher = "Elsevier",
    	volume = 484,
    	pages = "302--337",
    	issn = "0020-0255",
    	url = "http://www.sciencedirect.com/science/article/pii/S0020025519300945",
    	doi = "10.1016/j.ins.2019.01.076",
    	sm = "publications/supplementary_materials/SM_Parmezan_INS_2019_Evaluation.pdf",
    	address = "New York, United States of America",
    	month = "May",
    	year = 2019
    }
    
2.
  1. Huei Diana Lee, Ana Isabel Mendes, Newton Spolaôr, Jefferson Tales Oliva, Antonio Rafael Sabino Parmezan, Feng Chung Wu and Rui Manuel Fonseca-Pinto. Dermoscopic assisted diagnosis in melanoma: Reviewing results, optimizing methodologies and quantifying empirical guidelines. Knowledge-Based Systems 158:9–24, October 2018. URL, DOI BibTeX

    @article{Art:Lee:KNOSYS:2018:Dermoscopic,
    	author = "Lee, Huei Diana and Mendes, Ana Isabel and Spolaôr, Newton and Oliva, Jefferson Tales and Parmezan, Antonio Rafael Sabino and Wu, Feng Chung and Fonseca-Pinto, Rui Manuel",
    	title = "Dermoscopic assisted diagnosis in melanoma: Reviewing results, optimizing methodologies and quantifying empirical guidelines",
    	journal = "Knowledge-Based Systems",
    	publisher = "Elsevier",
    	volume = 158,
    	pages = "9--24",
    	issn = "0950-7051",
    	url = "http://www.sciencedirect.com/science/article/pii/S0950705118302454",
    	doi = "10.1016/j.knosys.2018.05.016",
    	address = "United States of America",
    	month = "Oct",
    	year = 2018
    }
    
3.
  1. Antonio Rafael Sabino Parmezan, Huei Diana Lee and Feng Chung Wu. Metalearning for choosing feature selection algorithms in data mining: Proposal of a new framework. Expert Systems with Applications 75:1–24, June 2017. URL, DOI, Supplementary Material BibTeX

    @article{Art:Parmezan:ESWA:2017:Metalearning,
    	author = "Parmezan, Antonio Rafael Sabino and Lee, Huei Diana and Wu, Feng Chung",
    	title = "Metalearning for choosing feature selection algorithms in data mining: Proposal of a new framework",
    	journal = "Expert Systems with Applications",
    	publisher = "Pergamon",
    	volume = 75,
    	pages = "1--24",
    	issn = "0957-4174",
    	url = "http://www.sciencedirect.com/science/article/pii/S0957417417300222",
    	doi = "10.1016/j.eswa.2017.01.013",
    	sm = "publications/supplementary_materials/SM_Parmezan_ESWA_2016_Metalearning.pdf",
    	address = "Tarrytown, United States of America",
    	month = "Jun",
    	year = 2017
    }
    
Book Chapters
1.
  1. Newton Spolaôr, Rui Manuel Fonseca-Pinto, Ana Isabel Mendes, Leandro Augusto Ensina, Weber Shoity Resende Takaki, Antonio Rafael Sabino Parmezan, Conceição Veloso Nogueira, Claudio Saddy Rodrigues Coy, Feng Chung Wu and Huei Diana Lee. Evaluating intelligent methods for decision making support in Dermoscopy based on information gain and ensemble (in press). In Witold Pedrycz, Luis Martinez, Rafael Espin and Gilberto Rivera (eds.). Computational Intelligence for Business Analytics. Studies in Computational Intelligence series, volume 1, Springer, 2021, pages 1–14. CIBA 2020. DOI BibTeX

    @incollection{Inc:Spolaor:CIBA:2021:Evaluating,
    	author = "Spola{\^o}r, Newton and Fonseca-Pinto, Rui Manuel and Mendes, Ana Isabel and Ensina, Leandro Augusto and Takaki, Weber Shoity Resende and Parmezan, Antonio Rafael Sabino and Nogueira, Concei{\c{c}}{\~a}o Veloso and Coy, Claudio Saddy Rodrigues and Wu, Feng Chung and Lee, Huei Diana",
    	title = "Evaluating intelligent methods for decision making support in Dermoscopy based on information gain and ensemble (in press)",
    	booktitle = "Computational Intelligence for Business Analytics",
    	series = "Studies in Computational Intelligence",
    	publisher = "Springer",
    	editor = "Pedrycz, Witold and Martinez, Luis and Espin, Rafael and Rivera, Gilberto",
    	volume = 1,
    	pages = "1--14",
    	isbn = "",
    	url = "",
    	doi = "",
    	address = "",
    	month = "",
    	year = 2021,
    	note = "CIBA 2020"
    }
    
2.
  1. Antonio Rafael Sabino Parmezan, Vinícius Mourão Alves Souza and Gustavo Enrique Almeida Prado Alves Batista. Towards Hierarchical Classification of Data Streams. In Ruben Vera-Rodriguez, Julian Fierrez and Aythami Morales (eds.). Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. Lecture Notes in Computer Science series, volume 11401, Springer, March 2019, pages 314–322. CIARP 2018. URL, DOI BibTeX

    @incollection{Inc:Parmezan:CIARP:2019:Towards,
    	author = "Parmezan, Antonio Rafael Sabino and Souza, Vinícius Mourão Alves and Batista, Gustavo Enrique Almeida Prado Alves",
    	title = "Towards Hierarchical Classification of Data Streams",
    	booktitle = "Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications",
    	series = "Lecture Notes in Computer Science",
    	publisher = "Springer",
    	editor = "Vera-Rodriguez, Ruben and Fierrez, Julian and Morales, Aythami",
    	volume = 11401,
    	pages = "314--322",
    	isbn = "978-3-030-13469-3",
    	url = "https://link.springer.com/chapter/10.1007%2F978-3-030-13469-3_37",
    	doi = "10.1007/978-3-030-13469-3_37",
    	address = "Cham, Switzerland",
    	month = "Mar",
    	year = 2019,
    	note = "CIARP 2018"
    }
    
Conference Publications (Full Papers)
1.
  1. Lucas Henrique Mantovani Jacintho, Tiago Pinho Silva, Antonio Rafael Sabino Parmezan and Gustavo Enrique Almeida Prado Alves Batista. Brazilian Presidential Elections: Analysing Voting Patterns in Time and Space Using a Simple Data Science Pipeline. In Anais do VIII Symposium on Knowledge Discovery, Mining and Learning. October 2020, 217–224. URL, DOI BibTeX

    @inproceedings{Inp:Jacintho:KDMILE:2020:Brazilian,
    	author = "Jacintho, Lucas Henrique Mantovani and Silva, Tiago Pinho and Parmezan, Antonio Rafael Sabino and Batista, Gustavo Enrique de Almeida Prado Alves",
    	title = "Brazilian Presidential Elections: Analysing Voting Patterns in Time and Space Using a Simple Data Science Pipeline",
    	booktitle = "Anais do VIII Symposium on Knowledge Discovery, Mining and Learning",
    	publisher = "SBC",
    	location = "Evento Online",
    	pages = "217--224",
    	issn = "0000-0000",
    	url = "https://sol.sbc.org.br/index.php/kdmile/article/view/11979",
    	doi = "10.5753/kdmile.2020.11979",
    	address = "Porto Alegre, Brasil",
    	month = "Oct",
    	year = 2020
    }
    
2.
  1. Antonio Rafael Sabino Parmezan, Diego Furtado Silva and Gustavo Enrique Almeida Prado Alves Batista. A combination of local approaches for hierarchical music genre classification. In Proceedings of the 21st International Society for Music Information Retrieval Conference. October 2020, 740–747. URL, DOI BibTeX

    @inproceedings{Inp:Parmezan:ISMIR:2020:Combination,
    	author = "Parmezan, Antonio Rafael Sabino and Silva, Diego Furtado and Batista, Gustavo Enrique de Almeida Prado Alves",
    	title = "A combination of local approaches for hierarchical music genre classification",
    	booktitle = "Proceedings of the 21st International Society for Music Information Retrieval Conference",
    	publisher = "ISMIR",
    	pages = "740--747",
    	isbn = "978-0-9813537-0-8",
    	url = "https://doi.org/10.5281/zenodo.4245540",
    	doi = "10.5281/zenodo.4245540",
    	address = "Montreal, Canada",
    	month = "Oct",
    	year = 2020
    }
    
3.
  1. Newton Spolaôr, Rui Manuel Fonseca-Pinto, Ana Isabel Mendes, Leandro Augusto Ensina, Narco Afonso Ravazzoli Maciejewski, Weber Shoity Resende Takaki, Antonio Rafael Sabino Parmezan, Conceição Veloso Nogueira, Claudio Saddy Rodrigues Coy, Feng Chung Wu and Huei Diana Lee. Feature extraction, selection and classification in Dermoscopy: an experimental comparison of intelligent methods to support decision making in medicine. In International Virtual Workshop of Business Analytics Eureka. May 2019, 1–10. PDF BibTeX

    @inproceedings{Inp:Spolaor:EUREKA:2019:Feature,
    	author = "Spola{\^o}r, Newton and Fonseca-Pinto, Rui Manuel and Mendes, Ana Isabel and Ensina, Leandro Augusto and Maciejewski, Narco Afonso Ravazzoli and Takaki, Weber Shoity Resende and Parmezan, Antonio Rafael Sabino and Nogueira, Concei{\c{c}}{\~a}o Veloso and Coy, Claudio Saddy Rodrigues and Wu, Feng Chung and Lee, Huei Diana",
    	title = "Feature extraction, selection and classification in Dermoscopy: an experimental comparison of intelligent methods to support decision making in medicine",
    	booktitle = "International Virtual Workshop of Business Analytics Eureka",
    	pages = "1--10",
    	pdf = "publications/pdf/Inp_Spolaor_EUREKA_2019_Feature.pdf",
    	address = "Ciudad Ju{\'a}rez, Mexico",
    	month = "May",
    	year = 2019
    }
    
4.
  1. Antonio Rafael Sabino Parmezan and Gustavo Enrique Almeida Prado Alves Batista. A study of the use of complexity measures in the similarity search process adopted by kNN algorithm for time series prediction. In 2015 IEEE 14th International Conference on Machine Learning and Applications (ICMLA). December 2015, 45–51. URL, DOI BibTeX

    @inproceedings{Inp:Parmezan:ICMLA:2015:Study,
    	author = "Antonio Rafael Sabino Parmezan and Gustavo Enrique Almeida Prado Alves Batista",
    	title = "A study of the use of complexity measures in the similarity search process adopted by {kNN} algorithm for time series prediction",
    	booktitle = "2015 IEEE 14th International Conference on Machine Learning and Applications (ICMLA)",
    	publisher = "IEEE",
    	pages = "45--51",
    	crossref = "DBLP:conf/icmla/2015",
    	url = "http://dx.doi.org/10.1109/ICMLA.2015.217",
    	doi = "10.1109/ICMLA.2015.217",
    	address = "Miami, United States of America",
    	month = "Dec",
    	year = 2015
    }
    
Conference Publications (Short Papers)
1.
  1. Newton Spolaôr, Huei Diana Lee, Weber Shoity Resende Takaki, Antonio Rafael Sabino Parmezan and Feng Chung Wu. Using label powerset for multi-label feature selection: an experimental comparison. In Eureka International Virtual Meeting. December 2016, 1–1. URL PDF BibTeX

    @inproceedings{Inp:Spolaor:EUREKA:2016:Using,
    	author = "Spola{\^o}r, Newton and Lee, Huei Diana and Takaki, Weber Shoity Resende and Parmezan, Antonio Rafael Sabino and Wu, Feng Chung",
    	title = "Using label powerset for multi-label feature selection: an experimental comparison",
    	booktitle = "Eureka International Virtual Meeting",
    	pages = "1--1",
    	url = "http://eurekaiberoamerica.net/567-2/",
    	pdf = "https://drive.google.com/file/d/0BzEwLzQRvu0gLWIyVXlVRVk4d1A3bDhwZW4wVGs0aVpPZ1pB/view",
    	address = "Torreon, Mexico",
    	month = "Dec",
    	year = 2016
    }
    
2.
  1. Antonio Rafael Sabino Parmezan, Feng Chung Wu and Huei Diana Lee. Meta-Aprendizado no Auxílio à Seleção de Atributos: Um Estudo para Medidas de Correlação e Consistência. In XXI Encontro Anual de Iniciação Científica (EAIC). October 2012, 1–4. URL PDF BibTeX

    @inproceedings{Inp:Parmezan:EAIC:2012:Meta_Aprendizado,
    	author = "Parmezan, Antonio Rafael Sabino and Wu, Feng Chung and Lee, Huei Diana",
    	title = "Meta-Aprendizado no Aux{\'i}lio {\`a} Sele{\c{c}}{\~a}o de Atributos: Um Estudo para Medidas de Correla{\c{c}}{\~a}o e Consist{\^e}ncia",
    	booktitle = "XXI Encontro Anual de Inicia{\c{c}}{\~a}o Cient{\'i}fica (EAIC)",
    	pages = "1--4",
    	url = "https://www.researchgate.net/publication/303875675",
    	pdf = "publications/pdf/Inp_Parmezan_EAIC_2012_Meta_Aprendizado.pdf",
    	address = "Maring\'a, Brasil",
    	month = "Oct",
    	year = 2012
    }
    
3.
  1. Antonio Rafael Sabino Parmezan, Huei Diana Lee and Feng Chung Wu. Estudo Preliminar da Construção de um Modelo de Recomendação de Algoritmos de Seleção de Atributos utilizando Meta-Aprendizado. In XX Simpósio Internacional de Iniciação Científica da Universidade de São Paulo (SIICUSP). October 2012, 1–1. URL PDF BibTeX

    @inproceedings{Inp:Parmezan:SIICUSP:2012:Estudo,
    	author = "Parmezan, Antonio Rafael Sabino and Lee, Huei Diana and Wu, Feng Chung",
    	title = "Estudo Preliminar da Constru{\c{c}}{\~a}o de um Modelo de Recomenda{\c{c}}{\~a}o de Algoritmos de Sele{\c{c}}{\~a}o de Atributos utilizando Meta-Aprendizado",
    	booktitle = "XX Simp{\'o}sio Internacional de Inicia{\c{c}}{\~a}o Cient{\'i}fica da Universidade de S{\~a}o Paulo (SIICUSP)",
    	pages = "1--1",
    	url = "https://www.researchgate.net/publication/303875673",
    	pdf = "publications/pdf/Inp_Parmezan_SIICUSP_2012_Estudo.pdf",
    	address = "S{\~a}o Paulo, Brasil",
    	month = "Oct",
    	year = 2012
    }
    
4.
  1. Antonio Rafael Sabino Parmezan, Huei Diana Lee and Feng Chung Wu. Redução da Dimensionalidade em Bases de Dados Naturais através de Métodos de Filtro para Seleção de Atributos Importantes. In XIX Simpósio Internacional de Iniciação Científica da Universidade de São Paulo (SIICUSP). November 2011, 1–1. URL PDF BibTeX

    @inproceedings{Inp:Parmezan:SIICUSP:2011:Reducao,
    	author = "Parmezan, Antonio Rafael Sabino and Lee, Huei Diana and Wu, Feng Chung",
    	title = "Redu{\c{c}}{\~a}o da Dimensionalidade em Bases de Dados Naturais atrav{\'e}s de M{\'e}todos de Filtro para Sele{\c{c}}{\~a}o de Atributos Importantes",
    	booktitle = "XIX Simp{\'o}sio Internacional de Inicia{\c{c}}{\~a}o Cient{\'i}fica da Universidade de S{\~a}o Paulo (SIICUSP)",
    	pages = "1--1",
    	url = "https://www.researchgate.net/publication/303875739",
    	pdf = "publications/pdf/Inp_Parmezan_SIICUSP_2011_Reducao.pdf",
    	address = "S{\~a}o Carlos, Brasil",
    	month = "Nov",
    	year = 2011
    }
    
5.
  1. Antonio Rafael Sabino Parmezan, Feng Chung Wu and Huei Diana Lee. Estudo de Medidas de Importância e Algoritmos para Seleção de Atributos para Mineração de Dados. In XX Encontro Anual de Iniciação Científica (EAIC). October 2011, 1–4. URL PDF BibTeX

    @inproceedings{Inp:Parmezan:EAIC:2011:Estudo,
    	author = "Parmezan, Antonio Rafael Sabino and Wu, Feng Chung and Lee, Huei Diana",
    	title = "Estudo de Medidas de Import{\^a}ncia e Algoritmos para Sele{\c{c}}{\~a}o de Atributos para Minera{\c{c}}{\~a}o de Dados",
    	booktitle = "XX Encontro Anual de Inicia{\c{c}}{\~a}o Cient{\'i}fica (EAIC)",
    	pages = "1--4",
    	url = "https://www.researchgate.net/publication/303875827",
    	pdf = "publications/pdf/Inp_Parmezan_EAIC_2011_Estudo.pdf",
    	address = "Ponta Grossa, Brasil",
    	month = "Oct",
    	year = 2011
    }
    
6.
  1. Antonio Rafael Sabino Parmezan, Huei Diana Lee, Carlos Andrés Ferrero, Willian Zalewski, André Gustavo Maletzke and Feng Chung Wu. Estudo Comparativo entre Métodos de Seleção de Atributos Baseados em Medidas de Precisão e Correlação Aplicados à Bases de Dados. In XVIII Simpósio Internacional de Iniciação Científica da Universidade de São Paulo (SIICUSP). November 2010, 1–1. URL PDF BibTeX

    @inproceedings{Inp:Parmezan:SIICUSP:2010:Estudo,
    	author = "Parmezan, Antonio Rafael Sabino and Lee, Huei Diana and Ferrero, Carlos Andr{\'e}s and Zalewski, Willian and Maletzke, Andr{\'e} Gustavo and Wu, Feng Chung",
    	title = "Estudo Comparativo entre M{\'e}todos de Sele{\c{c}}{\~a}o de Atributos Baseados em Medidas de Precis{\~a}o e Correla{\c{c}}{\~a}o Aplicados \`a Bases de Dados",
    	booktitle = "XVIII Simp{\'o}sio Internacional de Inicia{\c{c}}{\~a}o Cient{\'i}fica da Universidade de S{\~a}o Paulo (SIICUSP)",
    	pages = "1--1",
    	url = "https://www.researchgate.net/publication/303875731",
    	pdf = "publications/pdf/Inp_Parmezan_SIICUSP_2010_Estudo.pdf",
    	address = "S{\~a}o Paulo, Brasil",
    	month = "Nov",
    	year = 2010
    }
    
Technical Reports
1.
  1. Antonio Rafael Sabino Parmezan and Gustavo Enrique Almeida Prado Alves Batista. Descrição de Modelos Estatísticos e de Aprendizado de Máquina para Predição de Séries Temporais. Number 412, Relatórios Técnicos do ICMC-USP, Instituto de Ciências Matemáticas e de Computação, Universidade de São Paulo, São Carlos, 2016, August 2016. URL PDF BibTeX

    @techreport{Tech:Parmezan:USP:2016:Descricao,
    	author = "Antonio Rafael Sabino Parmezan and Gustavo Enrique Almeida Prado Alves Batista",
    	title = "Descri{\c{c}}{\~a}o de Modelos Estat\'isticos e de Aprendizado de M\'aquina para Predi{\c{c}}{\~a}o de S\'eries Temporais",
    	institution = "Relat{\'o}rios T{\'e}cnicos do ICMC-USP, Instituto de Ci\^encias Matem\'aticas e de Computa\c{c}\~ao, Universidade de S\~ao Paulo, S\~ao Carlos, 2016",
    	isbn = "0103-2569",
    	number = 412,
    	pages = "1--96",
    	url = "http://bdpi.usp.br/single.php?_id=002772986",
    	pdf = "publications/pdf/Tech_Parmezan_USP_2016_Descricao.pdf",
    	address = "S\~ao Carlos, Brasil",
    	month = "Aug",
    	year = 2016
    }
    
2.
  1. Antonio Rafael Sabino Parmezan, Huei Diana Lee, Newton Spolaôr and Feng Chung Wu. Avaliação de Métodos para Seleção de Atributos Importantes para Aprendizado de Máquina Supervisionado no Processo de Mineração de Dados. Number 2, Relatórios Técnicos do LABI-UNIOESTE, Laboratório de Bioinformática, Universidade Estadual do Oeste do Paraná, Foz do Iguaçu, 2012, December 2012. URL PDF BibTeX

    @techreport{Tech:Parmezan:UNIOESTE:2012:Avaliacao,
    	author = "Parmezan, Antonio Rafael Sabino and Lee, Huei Diana and Spola{\^o}r, Newton and Wu, Feng Chung",
    	title = "Avalia{\c{c}}{\~a}o de M{\'e}todos para Sele{\c{c}}{\~a}o de Atributos Importantes para Aprendizado de M{\'a}quina Supervisionado no Processo de Minera{\c{c}}{\~a}o de Dados",
    	institution = "Relat{\'o}rios T{\'e}cnicos do LABI-UNIOESTE, Laborat{\'o}rio de Bioinform{\'a}tica, Universidade Estadual do Oeste do Paran{\'a}, Foz do Igua{\c{c}}u, 2012",
    	number = 2,
    	pages = "1--58",
    	url = "https://www.researchgate.net/publication/303875686",
    	pdf = "publications/pdf/Tech_Parmezan_UNIOESTE_2012_Avaliacao.pdf",
    	address = "Foz do Igua{\c{c}}u, Brasil",
    	month = "Dec",
    	year = 2012
    }
    
Datasets
1.
  1. Antonio Rafael Sabino Parmezan and Gustavo Enrique Almeida Prado Alves Batista. ICMC-USP Time Series Prediction Repository. March 2014. Instituto de Ciências Matemáticas e de Computação, Universidade de São Paulo, São Carlos, 2014. URL PDF BibTeX

    @misc{Repository:Parmezan:USP:2014:TSPR,
    	author = "Antonio Rafael Sabino Parmezan and Gustavo Enrique Almeida Prado Alves Batista",
    	title = "{ICMC-USP} Time Series Prediction Repository",
    	note = "Instituto de Ci\^encias Matem\'aticas e de Computa\c{c}\~ao, Universidade de S\~ao Paulo, S\~ao Carlos, 2014",
    	url = "http://sites.labic.icmc.usp.br/icmc_tspr/",
    	pdf = "publications/pdf/Repository_Parmezan_USP_2014_TSPR.pdf",
    	address = "S\~ao Carlos, Brasil",
    	month = "Mar",
    	year = 2014
    }
    
M.Sc. Dissertation and B.Sc. Monograph
1.
  1. Antonio Rafael Sabino Parmezan. Predição de Séries Temporais por Similaridade. June 2016. Dissertação (Mestrado em Ciências de Computação e Matemática Computacional) - Instituto de Ciências Matemáticas e de Computação, Universidade de São Paulo, São Carlos, 2016. URL, DOI BibTeX

    @misc{MSc:Parmezan:UNIOESTE:2016:Proposta,
    	author = "Parmezan, Antonio Rafael Sabino",
    	title = "Predi{\c{c}}{\~a}o de S\'eries Temporais por Similaridade",
    	institution = "Universidade de S\~ao Paulo",
    	note = "Disserta{\c{c}}{\~a}o (Mestrado em Ci\^encias de Computa{\c{c}}{\~a}o e Matem\'atica Computacional) - Instituto de Ci\^encias Matem\'aticas e de Computa\c{c}\~ao, Universidade de S\~ao Paulo, S\~ao Carlos, 2016",
    	pages = "1--219",
    	url = "http://www.teses.usp.br/teses/disponiveis/55/55134/tde-21112016-150659",
    	doi = "10.11606/d.55.2016.tde-21112016-150659",
    	address = "S\~ao Carlos, Brasil",
    	month = "Jun",
    	year = 2016
    }
    
2.
  1. Antonio Rafael Sabino Parmezan. Proposta de um Módulo de Monitoramento de Qualidade de Dados com Assistência Inteligente: Um Estudo de Caso para o Sistema Médico de Auxílio à Cirurgia Coloproctológica. November 2012. Monografia (Graduação em Ciência da Computação) - Universidade Estadual do Oeste do Paraná, Foz do Iguaçu, 2012. URL PDF BibTeX

    @misc{BSc:Parmezan:UNIOESTE:2012:Proposta,
    	author = "Parmezan, Antonio Rafael Sabino",
    	title = "Proposta de um M{\'o}dulo de Monitoramento de Qualidade de Dados com Assist{\^e}ncia Inteligente: Um Estudo de Caso para o Sistema M{\'e}dico de Aux{\'i}lio {\`a} Cirurgia Coloproctol{\'o}gica",
    	institution = "Universidade Estadual do Oeste do Paran{\'a}",
    	note = "Monografia (Gradua{\c{c}}{\~a}o em Ci\^encia da Computa{\c{c}}{\~a}o) - Universidade Estadual do Oeste do Paran{\'a}, Foz do Igua{\c{c}}u, 2012",
    	pages = "1--100",
    	url = "https://www.researchgate.net/publication/303875852",
    	pdf = "publications/pdf/BSc_Parmezan_UNIOESTE_2012_Proposta.pdf",
    	address = "Foz do Igua{\c{c}}u, Brasil",
    	month = "Nov",
    	year = 2012
    }
    
Recent Funding
2020 - Present USAID Ph.D. Research Scholarship
2017 - 2020 CNPq Ph.D. Research Scholarship
2018 - 2019 PDSE/CAPES Ph.D. Sandwich Program Abroad
2016 - 2017 CNPq Research Technical Support Fellowship - Level 1A
2013 - 2015 FAPESP M.Sc. Research Scholarship
2011 - 2012 PIBIC/PRPPG/UNIOESTE B.Sc. Research Scholarship
2010 - 2011 PIBIC/CNPq/UNIOESTE B.Sc. Research Scholarship
Useful Links
Datasets for Data Mining

Research Groups
Miscellany
Follow a little bit of my daily life on my Instagram profile (@aparmezan). I love to read, draw, cook, work out, travel, and talk about my experiences. Spending time with my family and friends is what I love most.
Contact Information
Antonio Rafael Sabino Parmezan

Universidade de São Paulo - USP
Instituto de Ciências Matemáticas e de Computação - ICMC
Avenida Trabalhador São-carlense, 400 / Centro
ZIP code 13566-590 / São Carlos - SP, Brazil

Phone: +55-16-3373-9646
Email: parmezan at usp br