Advanced search
Start date

Supervised learning on computer-aided discrete response data with applications in imbalanced data


The main objective of this project is to propose, estimate and apply different supervised learning models for Discrete response with emphases in binary response. Specifically, the project aims to develop new classification models to the case of discrete responses, considering by example proposals of new links for binary regression models. Extensions this models for mixed regression model and latent variables as item response theory and cognitive diagnostic will be also considered. We focus in new estimation methods, including simulation studies and application studies to real data complete the objectives of this project. The proposal is justified by the shortage of research that accommodates such kind of data, the practical implications of the results of such modeling, for the relevance of the working together with international and national researches, develop orientations in students and apply the proposed methodologies in the context of data science and machine learning. It is expected to develop codes in R and Python as Computer-aided statistical data analysis for the different proposed models and then make it available to users in free repositories and provide data used in this research for replication of proposed methods, dissemination and development of models of the proposed models and to publish papers in relevant international journals, made presentations in scientific meetings to disseminate the results obtained. (AU)

Articles published in Agência FAPESP Newsletter about the research grant:
Articles published in other media outlets (0 total):
More itemsLess items

Scientific publications (6)
(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)
DE OLIVEIRA, EDUARDO S. B.; DE CASTRO, MARIO; BAYES, CRISTIAN L.; BAZAN, JORGE L.. Bayesian quantile regression models for heavy tailed bounded variables using the No-U-Turn sampler. Computational Statistics, v. N/A, p. 34-pg., . (21/11720-0)
ORDONEZ, JOSE A.; PRATES, MARCOS O.; BAZAN, JORGE L.; LACHOS, VICTOR H.. Penalized complexity priors for the skewness parameter of power links. CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE, v. N/A, p. 20-pg., . (21/11720-0)
ORDONEZ, JOSE A.; PRATES, MARCOS O.; MATOS, LARISSA A.; LACHOS, VICTOR H.. Objective Bayesian analysis for geostatistical Student-t processes. JOURNAL OF SPATIAL SCIENCE, v. N/A, p. 19-pg., . (21/11720-0)
LACHOS, VICTOR H.; BAZAN, JORGE L.; CASTRO, LUIS M.; PARK, JIWON. The skew-t censored regression model: parameter estimation via an EM-type algorithm. COMMUNICATIONS FOR STATISTICAL APPLICATIONS AND METHODS, v. 29, n. 3, p. 19-pg., . (21/11720-0)
COELHO, FABIANO R.; RUSSO, CIBELE M.; BAZAN, JORGE L.. On outliers detection and prior distribution sensitivity in standard skew-probit regression models. BRAZILIAN JOURNAL OF PROBABILITY AND STATISTICS, v. 36, n. 3, p. 22-pg., . (21/11720-0)
BAZAN, JORGE LUIS; ARI, SANDRA ELIZABETH FLORES; AZEVEDO, CAIO L. N.; DEY, DIPAK K.. Revisiting the Samejima-Bolfarine-Bazan IRT models: New features and extensions. BRAZILIAN JOURNAL OF PROBABILITY AND STATISTICS, v. 37, n. 1, p. 25-pg., . (21/11720-0)

Please report errors in scientific publications list using this form.