Fscores_maker: Calculate F_1 scores based on SVM evaluations

In martigso/tonR: Tools for operating the Talk of Norway dataset in R

Description

A function for calculating F_1 scores based on SVM evaluations

Usage

Fscores_maker(actual, pred)

Arguments

data	Data frame containing the actual and predicted values of the property under consideration
actual	Character string of the column name of the the actual values of the property under consideration
pred	Character string of the column name of the the predicted values of the property under consideration. Must be same length as actual.
parties	Logical vector specifying if actual and pred is party labels

Details

The F-scores are calculated as a product of true positives (TP), false negatives (FN), and false positives (FP).

Precision:

P = TP / (TP + FP)

Recall:

R = TP / (TP + FN)

F-score:

F = 2 * ((P * R) / (P + R))

Value

Returns a list with a data frame of F-scores, precision, and recall for each class, macro averaged F-score, and accuracy.

See Also

svm Fscores_boot

Other fscores: Fscores_boot

Examples

data(tonDemo)
Fscores_maker(tonDemo, "party_id", "class_token")

martigso/tonR documentation built on May 21, 2019, 12:38 p.m.