Date & time: Thursday 22 September 2016, 13.30-17.00
Objective: Provide demo of functionality of the GSIF and plotKML packages with a focus on automated mapping
General description: plotKML (R package) is a platform for scientific visualization of spatio-temporal data and models. It allows fast parsing of spatial, spatio-temporal and profile objects in R to KML and allows users to quickly visualize results of spatial analysis from R to Google Earth. GSIF package contains wrapper functions for automated soil mapping in 2D and 3D; current version of the autopredict function relies on using Machine Learning algorithms mainly from ranger and quantregforest (quantile regression forests) packages. By combining GSIF and the plotKML packages spatial predictions can be created and visualized in Google Earth by using few lines of code.
Required back-ground knowledge: gstat, sp and spacetime classes in R; machine learning, variogram modelling;
Software / R packages required: plotKML, GSIF, sp, raster, rgdal, gstat, ranger, randomForestSRC, xgboost, aqp, XML, randomForest, quantregForest;
Software installation instructions:
12:30–13:30 Lunch break
13:30–15:00 Installation of the package and general functionality; using machine learning to create spatial predictions:
15:30–17:00 Importing and visualizing profile data, polygon maps and gridded maps; making visualizations with your own datasets (1.5 hrs)
- Machine learning using soil data (tutorial)
- Hengl, T., Roudier, P., Beaudette, D., & Pebesma, E. (2015). plotKML: scientific visualization of spatio-temporal data. Journal of Statistical Software, 63(5).
- Yau, N. 2011. Visualize this: The FlowingData Guide to Design, Visualization, and Statistics. Wiley
- Wernecke, J. 2010: The KML handbook: geographic visualization for the Web.