MelbURN is excited to be hosting our own mini version of the useR! conference! For the first time ever, the annual useR! conference is being held down under (in Brisbane), and we're privileged to have two of the speakers presenting to the local Melbourne R community.
PS for those thinking of making the trip to Brisy, check out the full useR! program: https://user2018.r-project.org/
We're also excited to have the Amazon Melbourne Office as a new venue for our meetups - many thanks to Amazon for their support!
* Please arrive before 6pm which is when access to the lifts becomes restricted *
5:45 Pizzas, drinks & networking
6:30 First talk: Learning from (dis)similarity data
7:00 Short break
7:10 Second talk: Interactive Graphics for Visually Diagnosing Forest Classifiers in R
7:40 More networking
First talk: Nathalie Vialaneix - Learning from (dis)similarity data
In some applications and in order to better address real-world
situations, data can be known through pairwise measures of resemblance
or difference between the objects of interest (similarities,
dissimilarities, kernels, networks...). This talk will describe a
general framework to deal with such data, especially focusing on the
unsupervised setting and exploratory analyses. Also, solutions for
combining multiple relational data - each providing a different view on
a specific aspect of the data - will be described. The talk will provide
an overview of applications of this framework to self-organizing maps (R
package SOMbrero), constrained hierarchical clustering (R package
adjclust) and PCA (R package mixKernel), with illustrations on case
studies in the fields of biology and social sciences.
Nathalie is a researcher at the French National Institute for
Agronomical Research (INRA) in the Unit of Applied Mathematics and
Computer Sciences in Toulouse. She received her PhD in Mathematics from the University Toulouse 2 (Le Mirail), in 2005. She is a board member of the biostatistics platform in Toulouse and a former board member of the French Statistical Association (SFdS). Her research interests include
machine learning and network analysis with applications to social sciences or omics data. She is the maintainer of the SOMbrero, SISIR and
RNAseqNet R packages and author of a number of others.
Second talk: Natalia da Silva - Interactive Graphics for Visually Diagnosing Forest Classifiers in R
This research describes structuring data and constructing plots to explore forest classification models interactively. A forest classifier is an example of an ensemble since it is produced by bagging multiple trees. The process of bagging and combining results from multiple trees produces numerous diagnostics which, with interactive graphics, can provide a lot of insight into class structure in high dimensions. Various aspects are explored in this paper, to assess model complexity, individual model contributions, variable importance and dimension reduction, and uncertainty in prediction associated with individual observations. The ideas are applied to the random forest algorithm and projection pursuit forest, but could be more broadly applied to other bagged ensembles. Interactive graphics are built in R using the ggplot2, plotly and shiny packages.
About the speaker: I’m an Assistant Professor in the Department of Statistics at the Universidad de la República in Montevideo, Uruguay. I obtained my Ph.D. degree in Statistics at Iowa State University on July 2017 working with Di Cook and Heike Hofmann. My interest are: supervised learning methods, prediction, exploratory data analysis, statistical graphics, reproducible research and meta-analysis. I’m co-founder of R-Ladies-Ames and R-Ladies-Montevideo. I’m working in different initiatives to get a stronger and bigger R community in Latin America.