Description Usage Arguments Value References Examples
Get accuracy, precision, and recall for multi-class, multi-tag, predictions.
1 | evaluate(x, known)
|
x |
The model classification |
known |
The known expert coded |
Returns a list of seven elements:
N |
The number of elements being assessed |
confusion_matrix |
A list of confusion matrices for each tag |
tag_accuracy |
A tag named vector of accuracies computed from the confusion matrices; (tp + tn)/(tp + tn + fp + fn) |
tag_precision |
A tag named vector of precisions computed from the confusion matrices; tp/(tp + fp) |
tag_recall |
A tag named vector of accuracies computed from the confusion matrices; tp/(tp + fn) |
macro_averaged |
Macro averaged accuracy, precision, and recall; computed accuracy, precision, and recall for each confusion matrix and average |
micro_averaged |
Micro averaged accuracy, precision, and recall; add the confusion amtrices and compute accuracy, precision, and recall |
https://www.youtube.com/watch?v=OwwdYHWRB5E&index=31&list=PL6397E4B26D00A269
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 | known <- list(1:3, 3, NA, 4:5, 2:4, 5, integer(0))
tagged <- list(1:3, 3, 4, 5:4, c(2, 4:3), 5, integer(0))
evaluate(tagged, known)
## Examples
library(dplyr)
data(presidential_debates_2012)
discoure_markers <- list(
response_cries = c("\\boh", "\\bah", "\\baha", "\\bouch", "yuk"),
back_channels = c("uh[- ]huh", "uhuh", "yeah"),
summons = "hey",
justification = "because"
)
## Only Single Tag Allowed Per Text Element
mod1 <- presidential_debates_2012 %>%
with(., term_count(dialogue, TRUE, discoure_markers)) %>%
classify()
fake_known <- mod1
set.seed(1)
fake_known[sample(1:length(fake_known), 300)] <- "random noise"
evaluate(mod1, fake_known)
## Multiple Tags Allowed
mod2 <- presidential_debates_2012 %>%
with(., term_count(dialogue, TRUE, discoure_markers)) %>%
classify(n = 2)
fake_known2 <- mod2
set.seed(30)
fake_known2[sample(1:length(fake_known2), 500)] <- c("random noise", "back_channels")
(myacc <- evaluate(mod2, fake_known2))
myacc$confusion_matrix
myacc$tag_accuracy
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.