I’m a statistician, reproducible research enthusiast,
git active user. I like statistical modelling
in general, package and dashboards development, text mining and
APIs. Also, I’m specially interested in Bayesian inference,
multivariate analysis, feature extraction and
probabilistic graphical models.
I’m a member of the R-Ladies, a non-profit organization that promotes gender minorities in the R/data science community. I’m the organiser of the R-Ladies Curitiba chapter, being also involved with the R-Ladies São Paulo and R-Ladies Dublin groups.
Being very interested in Music Information Retrieval, I have
developed, so far, two packages for music data extraction in
R: vagalumeR, which is about
getting lyrics data from the Vagalume API,
and chorrrds, a package that
extracts music chords from the CifraClub website
(both available on CRAN). I have contributed to the
R package that
makes the connection with the Spotify API
(found at this repository).
More about music data extraction & analysis can be found at the
R-Music blog, a blog that I started in 2018
and that has as its main goals to enable the study and practice of Music
Information Retrieval (MIR) in R.
Some of my work with Shiny Applications, and the
package can be found here
and here. The
first one is a non-linear models catalog, which is still
under development. The second is dedicated to the study
probability distributions in the unit interval, as it is also a catalog
on this subject.
Currently, I’m pursuing my PhD at the Hamilton Institute in Maynooth, Ireland. I’m a part of Andrew Parnell’s research group, and our work is mostly related to the development of new Bayesian methods for machine learning and its many possible applications.
Current PhD Candidate in Statistics - Bayesian Machine Learning, 2018
Hamilton Institute, Maynooth University
Bachelor in Statistics, 2018
Federal University of Paraná, Brazil
Tue, Jul 16, 2019, Machine Learning Summer School 2019
Mon, Jun 24, 2019,
Thu, May 23, 2019, IV International Seminar on Statistics with R
Tue, May 21, 2019, IV International Seminar on Statistics with R
Thu, May 16, 2019, 39th Conference on Applied Statistics in Ireland
rstudio::conf, Austin, US, January 2019
Tidyverse developer day, Austin, US, January 2019
XVI School of Regression Models, Pirenopólis, Goiás, Brazil, March 2019. Talk: Machine Learning And Chord Based Feature Engineering For Genre Prediction In Popular Brazilian Music
Women in Data Science Conference 2019 Zurich, Switzerland, April 2019.
APTS Courses week 2, Southampton, England, April 2019:
39th Conference On Applied Statistics In Ireland, Dundalk, Ireland, May 2019. Talk: Regularization Methods in Random Forests
IV International Seminar on Statistics with R, Niterói, Brazil, May 2019. Talk:
APTS Courses week 3, Durham, England, July 2019
Machine Learning Summer School, London, England, July 2019.
StanCon, Cambridge, England, August 2019
APTS Courses week 4, Oxford, England, September 2019
ISMIR 2019, Delft, The Netherlands, November 2019
R-Ladies is a worldwide organization whose mission it to promote gender diversity in the R comunnity.
Music information retrieval is the field that develops and applies computational tools combined with music theory. The area intends to amplify the understanding and utility of music data. With that in mind, the goal of this project is to provide the implementation to many different music information retrival techniques in R. The topics involve since packages for music data extraction, tools for data exploration and probabalistic modelling, including deep learning.
The goal of this project is to develop a complete framework in R for the use of non linear models. It includes a dashboard for the interactive visualization of all the present models, along with other functions, also useful for the fit of these models.
The main goal of this project is get together students, teachers, professionals and researchers interested in the study and building of Shiny applications. The motivation comes from the understanding that the learning of statistics, mathematics and other subjects can be facilitated with the use of interactive tools, such as Shiny applications.
Currently, I have two packages on CRAN
They have pretty much the same subject: music data.
vagalumeR is the package that
extracts music lyrics from the Vagalume
API. The second package,
chorrrds, extracts music chords from the CifraClub website, through web scraping
Both these packages are in constantly development, being that their new versions always come out first in the R-Music organization on GitHub.