The key components of dsos
depend on several publicly
available R
packages, including:
data.table
package (Dowle
and Srinivasan 2020) for faster data.frame
.ranger
package (Wright and
Ziegler 2017) for random forest;isotree
package (Cortes
2020) for isolation-based outlier detection;ggplot2
package (Wickham
2016) for visualizations;WeightedROC
package (Hocking
2020) for computing the test statistic;simctest
package (Gandy
2009) for sequential Monte Carlo tests;We are grateful to the authors of each of these packages for making their software freely available.