France-Mexico school on Data-Analysis

Instituto de Matemáticas de la UNAM, 
Ciudad Universitaria, CDMX, December 13-15, 2017

Consisting of three 2h courses and practical sessions: 

Avner Bar-Hen :

Modeling high dimensional data

The development of high-dimensional models is an ongoing challenge, 
especially when the number of covariates is much larger than the number of observations. 
We review classical aproaches and illustrate them through examples.

Xavier Gendre

Introduction to Model Selection

To handle data sets, we commonly need to choose a way to represent the data in a suitable space called a model. 
Such a choice is not innocent and the statistician has to pay attention because this might cause further difficulties regarding what we can say about a data analysis. 
We give an introduction to model selection by focusing on important intrinsic statistical concepts and illustrative examples.

Maguelonne Teisseire 

Spatial Text Mining

Technological advances in terms of data acquisition enable one to better monitor dynamic phenomena in various domains (areas, fields) including the environment. 
The collected data is more and more complex - spatial, temporal, heterogeneous and multi-scale. 
Exploiting this data requires new data analysis and knowledge discovery methods.
In that context, approaches aimed at extracting spatio-temporal information for text matching are particularly relevant. 
This course focuses on the identification of spatial information in document for automatic annotation and document matching.

Organized by: 

- Embajada de Francia en México
- Instituto de Matemáticas de la UNAM
- Institut Fourier, Université Grenoble-Alpes, France
- Laboratorio Solomon Lefschetz (LAISLA) CoNaCyT-CNRS

Eric Bonnetier
José Seade
Gerónimo Uribe Bravo