Projects

PonteItalia (Live Demo)

A web application to find genealogical information using public civil records from Italy. It is mostly used by people who are trying to find information about their ancestors for the purpose of double-citizenship. It is currently used by more than 5k people per month.

Face2Data (Code, 110⭐)

A Flask based application to retrieve information from a person based on its face. It uses a pre-trained Neural Network model to extract facial features and then predicts the gender, age and ethniticy of the person.

Embeddings for Categorical Variables (Code, 70⭐)

A Python library to generate embeddings for categorical variables using Deep Neural Networks.

DigiPathos (Code, 39 ⭐)

A CLI tool to retrieve plant diseases datasets from Brazilian Agricultural website. It is used by researchers and students to retrieve datasets for their researches.

PADify (Code, 39 ⭐)

A Python library to identify forged facial biometric samples with Neural Networks. It is used by researchers and students to identify forged facial biometric samples. This work was presented at the Conference of "Advanced Concepts for Intelligent Vision Systems (ACIVS), in New Zealand". Paper can be found here.

Research & Publications

Machine Learning to Predict Neonatal Mortality Using Public Health Data from São Paulo - Brazil (Paper)

European Population Conference (2020, Italy)

Maternal characteristics and the risk of neonatal mortality in Brazil between 2006 and 2016 (Paper)

International Journal of Population Studies

Determinants of the Neonatal Mortality Risk in Brazil, 2006-2016 - A Machine Learning Models Approach (Paper)

Population Association of America (2020, Washington DC)

Applied Machine Learning for Neonatal Mortality Risk Assessment - A Case Study Using Public Health Data from São Paulo (Paper)