build: Automatic build of the anomaly detection model

In jaroslav-kuchar/fpmoutliers: Frequent Pattern Mining Outliers

Description

An experimental implementataion that automatically builds an anomaly detection model

Usage

build(data, func = FPI, initial_support = 0.5, top_outlier_threshold = 3,
  iteration_timeout = 10)

Arguments

data	data.frame or transactions from arules with input data
func	function name of the method that will be use during the automatic build
initial_support	initial maximum support
top_outlier_threshold	number of top unique outliers as a stopping condition
iteration_timeout	timeout of one iteration

Value

model of outlier detection

Examples

# simple build with default parameters (FPI method)
library("fpmoutliers")
data("iris")
model <- fpmoutliers::build(iris[sample(nrow(iris), 5),])


# using other anomaly detection methods for the automatic build (e.g. LFPOF)
library("fpmoutliers")
data("iris")
model <- fpmoutliers::build(iris[sample(nrow(iris), 5),], func=LFPOF)

jaroslav-kuchar/fpmoutliers documentation built on May 18, 2019, 4:48 p.m.