GetLabels function

Get Lables

GetLabels Calculates the start and end positions of each window that are focused on the real anomalies. This windows can be used to know if the detected anomaly is a true positive or not.

GetLabels(data)

Arguments

• data: All dataset with training and test datasets with at least timestamp, value, is.anomaly, is.real.anomaly, start.limit and end.limit columns.

Returns

Same data set with two additional columns label and first.tp. first.tp indicates for each window Which is the position of first true positive. label indicates for each detection if it is a TP, FP, TN or FN.

Details

data must be a data.frame with timestamp, value, is.anomaly

and is.real.anomaly columns. timestamp column can be numeric, of type POSIXct, or a character type date convertible to POSIXct. see GetWindowsLimits to know more about how to get start.limit and end.limit columns.

Examples

## Generate data
set.seed(100)
n <- 180
x <- sample(1:100, n, replace = TRUE)
x[70:90] <- sample(110:115, 21, replace = TRUE)
x[25] <- 200
x[150] <- 170
df <- data.frame(timestamp = 1:n, value = x)

# Add is.real.anomaly column
df$is.real.anomaly <- 0
df[c(25,80,150), "is.real.anomaly"] <- 1

## Calculate anomalies
result <- CpSdEwma(
  data = df$value,
  n.train = 5,
  threshold = 0.01,
  l = 3
)
res <- cbind(df, result)

# Get Window Limits
data <- GetWindowsLimits(res)
data[data$is.real.anomaly == 1,]

# Get labels
data <- GetLabels(data)
data[data$is.real.anomaly == 1 | data$is.anomaly == 1,]

# Plot results
PlotDetections(res, print.real.anomaly = TRUE, print.time.window = TRUE)

References

A. Lavin and S. Ahmad, “Evaluating Real-time Anomaly Detection Algorithms – the Numenta Anomaly Benchmark,” in 14th International Conference on Machine Learning and Applications (IEEE ICMLA’15), 2015.