My research is focused on optimization techniques applied to data science problems, such as clustering, classification, regression and pattern/anomaly detection. My doctorate at the Computer Science department of PUC-Rio focused on semi-supervised learning in the presence of noisy and scarce information, along with optimization models and methods for clustering and graph partitioning problems.
Fields of interest. Optimization, Machine Learning, Semi-Supervised Learning, Data Visualization, Pattern Recognition, Mathematical Programming, Probability Theory, Clustering, Community Detection.
PhD in Computer Science (2021). A model-based framework for semi-supervised clustering and community detection.
Pontifical Catholic University of Rio de Janeiro (PUC-Rio), Brazil; with one year at Polytechnique Montréal, Canada, as a visiting student.
Master's in Computer Science (2017). Hybrid genetic algorithm for the minimum sum-of-squares clustering problem.
Pontifical Catholic University of Rio de Janeiro (PUC-Rio), Brazil.
Bachelor's in Information Systems (2014). A clustering-based approach to detect probable outcomes of lawsuits.
Federal University of the State of Rio de Janeiro (Unirio), Brazil.
Vector Trading, LLC: Quantitative developer (since Jul ' 2021)
Quantitative analysis, software development, R&D.
CIRRELT: Research visitor / Doctorate exchange (Oct ' 2018 — Oct ' 2019)
One-year scholarship at the Interuniversity Research Centre on Enterprise Networks, Logistics and Transportation (CIRRELT), as part of my doctorate research on semi-supervised clustering (CAPES funding, under the supervision of Michel Gendreau).
IBM Research Center: Research & Development intern (Jul ' 2012 — Jun ' 2014)
Software developer in the Social Data Analytics group, working in research projects involving machine
learning and data analytics, including natural language processing, sentiment analysis, data mining and visualization. Projects: Social network simulator applied for the US presidential race (2012);
Real-time sentiment analysis in social media: the FIFA 2014 World Cup case (2013-14); Data mining algorithms for event detection in social networks (2014).
Argo Internet: Web development intern (Feb ' 2009 — Jul ' 2011)
Project of interfaces, software technical specification, web development.
Semi-supervised clustering with inaccurate pairwise annotations. D. Gribel, M. Gendreau, and T. Vidal. Information Sciences 607, 441-457, 2022.
Assortative-constrained stochastic block models. D. Gribel, T. Vidal, and M. Gendreau. 25th International Conference on Pattern Recognition (ICPR). IEEE, 2021.
HG-means: A scalable hybrid genetic algorithm for minimum sum-of-squares clustering. D. Gribel and T. Vidal. Pattern Recognition 88, 569-583, 2019.
Separable convex optimization with nested lower and upper constraints. T. Vidal, D. Gribel, and P. Jaillet. INFORMS Journal on Optimization 1 (1), 71-90, 2018.
A clustering-based approach to detect probable outcomes of lawsuits. D. Gribel, M. Gatti, and L. Azevedo. 24th ACM International on Conference on Information and Knowledge Management (CIKM 2015), p. 1831-1834. Melbourne, 2015.
Large-scale multi-agent-based modeling and simulation of microblogging-based online social network. M. Gatti, A. P. Appel, C. Pinhanez, C. Santos, D. Gribel, P. Cavalin, and S. Barbosa Neto. 14th International Workshop on Multi-Agent-based Simulation (MABS 13). Saint Paul, Minnesota, 2013.
Visual analytics of sentiment trends in social media streams: The 2013 Confederation Cup case. M. Gatti, A. Rademaker, D. Lemes, P. Cavalin, C. Pinhanez, and R. de Paula. INFOVIS 2013. Atlanta, 2013.