Visual Integration of Data and Models in Classification


Bruno Schneider


Praia de Botafogo, 190 - sala 537


7 de Junho de 2018 às 16h

There is a growing interest in the field of machine learning, which enables a computer to learn from data instead of explicitly programming it to execute a particular task. Classification is one of the problems in machine learning. Examples go from automatically recognizing images after training a model with known instances, supporting the diagnosis of diseases using medical records, or categorizing our e-mail messages as spam or not spam. My research proposes an interactive visual setup for the exploration of classification results in tight integration with scalable model space representations. The goal is to give direct access to models and data spaces in classification, thus enabling the user to explore the relationships between these spaces and seek for classification patterns that are not visible through standard model performance metrics. So far, I focused on ensemble learning, in which several classifier models work together, and at the end, the final classification is a combination of the individual results. In this presentation, I will also show my current work on how data visualization can support the better understanding of errors in classification, in particular, the ones caused by a lack of similar training examples.

*Texto informado pelo autor. 


Bruno Schneider is a Ph.D. Candidate in the Data Analysis and Visualization Group (DBVIS), at the University of Konstanz, Germany. The head of the group is Prof. Dr. Daniel Keim, who has a distinguished career in the fields of Information Visualization and Visual Analytics. The research in his group focuses on identifying opportunities of connecting human domain expert knowledge with fully automated algorithms and processes in computer science, through interactive visual interfaces. Before moving to Germany, Bruno earned his Master degree in Rio de Janeiro, Brazil, from the School of Applied Math, FGV (2014).