anomalyDetection
anomalyDetection copied to clipboard
An R package for implementing augmented network log anomaly detection procedures
anomalyDetection 
anomalyDetection
implements procedures to aid in detecting network log anomalies. By combining various multivariate analytic approaches relevant to network anomaly detection, it provides cyber analysts efficient means to detect suspected anomalies requiring further evaluation.
Installation
You can install anomalyDetection
two ways.
- Using the latest released version from CRAN:
install.packages("anomalyDetection")
- Using the latest development version from GitHub:
if (packageVersion("devtools") < 1.6) {
install.packages("devtools")
}
devtools::install_github("koalaverse/anomalyDetection", build_vignettes = TRUE)
Learning
To get started with anomalyDetection
, read the intro vignette: vignette("Introduction", package = "anomalyDetection")
. This will provide a thorough introduction to the functions provided in the package.
References
Gutierrez, R.J., Boehmke, B.C., Bauer, K.W., Saie, C.M. & Bihl, T.J. (2017) "anomalyDetection
: Implementation of augmented network log anomaly detection procedures." The R Journal, 9(2), 354-365. link