FLUSS - Prediction score calculation

fluss_score(gtruth, extracted, data_size)

Arguments

gtruth

an int or vector of int with the ground truth index of segments.

extracted

an int or vector of int with the extracted indexes from fluss_extract().

data_size

an int. Size of original input data.

Value

Returns the score of predicted semantic transitions compared with the ground truth. Zero is the best, One is the worst.

References

  • Gharghabi S, Ding Y, Yeh C-CM, Kamgar K, Ulanova L, Keogh E. Matrix Profile VIII: Domain Agnostic Online Semantic Segmentation at Superhuman Performance Levels. In: 2017 IEEE International Conference on Data Mining (ICDM). IEEE; 2017. p. 117-26.

Website: https://sites.google.com/site/onlinesemanticsegmentation/

Website: http://www.cs.ucr.edu/~eamonn/MatrixProfile.html

See also

Other Semantic Segmentations: floss_cac(), floss_extract(), floss(), fluss_cac(), fluss_extract(), fluss()

Examples

data <- mp_fluss_data$tilt_abp$data[1:1000]
w <- 10
truth <- c(945, 875)
mp <- tsmp(data, window_size = w, verbose = 0)
mp <- fluss_cac(mp)
#> Error in fluss_cac(mp): First argument must be an object of class `MatrixProfile`.
mp <- fluss_extract(mp, 2)
#> Error in fluss_extract(mp, 2): First argument must be an object of class `ArcCount`.
score <- fluss_score(truth, mp$fluss, length(data))
#> Error in if (abs(extracted[i] - gtruth[j]) < abs(minv[j])) {    minv[j] <- abs(extracted[i] - gtruth[j])}: argument is of length zero