Introduction to a new open source R package - "spm: Spatial Predictive Modelling", using novel, accurate, hybrid geostatistical and machine learning methods
29 November 2017
This presentation is about a recently released package, "spm: Spatial Predictive Modelling", on CRAN for R users. It will briefly introduce spm and spatial predictive modeling, followed by the introduction to the developmental history of spm in:
- spatial predictive methods,
- new hybrid methods of geostatistical and machine learning methods,
- assessment of predictive accuracy,
- applications of spatial predictive models, and
- its functions.
It will then demonstrate how to apply some functions in spm to relevant datasets and to show the resultant improvement in predictive accuracy and modelling efficiency. Finally, feedback on various aspects is welcomed for future release.
Dr Jin Li has research experience in ecology, ecological and environmental modelling, spatial predictive modelling and statistical computing. As a scientist, Jin worked in Chinese Academy of Sciences, UNE and CSIRO. Jin joined Geoscience Australia in 2007 as a Spatial Modeller/Computational Statistician. Jin has 80+ various publications and 40+ presentations, with 7,000+ citations (ResearchGate). Jin was an Associate Editor and is an editorial board member of Acta Oecologica.
This talk is presented as part of the Wednesday Seminar Series.
Location: Sir Harold Raggatt Theatre, Geoscience Australia
Cost/bookings: Free, No bookings required
More information: email@example.com