Skip to content

Instantly share code, notes, and snippets.

@aaizemberg
Last active December 25, 2015 18:59
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save aaizemberg/7024913 to your computer and use it in GitHub Desktop.
Save aaizemberg/7024913 to your computer and use it in GitHub Desktop.
data visualization @ CISL 2013
topics
data scraping (http://cippec.org/data/ --> view source --> ...)
open data / okfn-ar (indec / registro para bajar los datos)
data mining
reducción de dimensiones
clustering (kmeans)
data visualization
thematic maps --> http://en.wikipedia.org/wiki/Thematic_map
Inspiración
===========
http://www.macrofocus.com/public/products/infoscope/infoscope.png
scatterplot n-dims --> 2
Dimensional Reduction for Data Mapping
http://www.r-project.org/doc/Rnews/Rnews_2003-3.pdf
los datos
=========
Andy Tow ( https://twitter.com/andy_tow )
http://cippec.org/data/#
indec
http://www.opex.sig.indec.gov.ar/codgeo/index.php?pagina=descargas
provincias, departamentos y localidades
desde R
=======
help(read.table)
# leer un archivo separado por TABs
mydata <- read.delim("C:\\Users\\ariel\\Documents\\2013\\CISL\\cippec_data_id.txt", row.names=1)
# explicar el concepto del algoritmo de clustering (en el pizarrón)
# ejecutar un proceso de clustering
cl <- kmeans(mydata, 5)
# cluster sin escalar/normalizar
cl$cluster
6 10 22 26 2 14 18 30 34 38 42 46 50 54 58 62 66 70 74 78 82 86 94 90
2 1 3 4 5 5 3 3 4 4 1 1 3 3 4 4 3 4 1 1 5 4 1 3
cl$size
[1] 6 1 7 7 3
#cluster escalando
mydata <- scale(mydata)
cl <- kmeans(mydata, 5)
cl$cluster
cl$cluster
6 10 22 26 2 14 18 30 34 38 42 46 50 54 58 62 66 70 74 78 82 86 94 90
2 5 1 3 4 2 5 5 1 5 3 5 2 5 3 3 5 5 5 3 2 5 4 5
cl$size
[1] 2 4 5 2 11
LLevar el resultado al QGIS
juntar un mapa con una tabla de datos
hacer un mapa tematico
FIN
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment