Abstract Detail

Conference Wide

Shipunov, Alexey [1].

Machine learning with R for botanists.

My workshop is about R, the great and most developed free statistical tool. R is extremely popular in science, but learning curve is low. I provide the workshop which intends to overview and teach advanced machine learning with R. This will include data mining, multivariate, and deep learning methods useful forbotanists. We will learn the very basics of multivariate plotting, including 3D and trellis approaches; non-supervised methods including principal component analysis (PCA) and its variants (like CCA), t-SNE, self-organizing maps (SOM), various clustering techiques including k-means, DBSCAN and mean-shift, and othermanifold and layout methods like deep maps. Then we will overview various supervised methods like recursive trees, bagging (RanfomForest) and boosting ensemble learning, proximity learning (kNN and others), and blackbox learning like support vector machines (SVM) and famous neural networks. Finally, we will learn selected semi-supervised methods. There will be also an opportunity to discuss your own data. Please note that I require you to have (1) R installed on your laptop (I prefer to work in basic R, without IDE; all operation systems are OK though) and (2) basic knowledge of R such as data loading,modifications, plotting, plus simple statistical tests. For beginners in R, there is another workshop available.

Related Links:
Visual Statistics. Use R!

1 - Minot State University, Biology, 500 University Ave, Minot, ND, 58707, United States

none specified

Presentation Type: Workshop
Abstract ID:48
Candidate for Awards:None

Copyright © 2000-2019, Botanical Society of America. All rights reserved