Calculates the proportion of correct predictions (True Positives + True Negatives) over all cases from a confusion matrix object, providing a measure of the classifier's overall correctness.
Arguments
- cm
A dx_cm object created by
dx_cm()
.- detail
Character specifying the level of detail in the output: "simple" for raw estimate, "full" for detailed estimate including 95% confidence intervals.
- ...
Additional arguments to pass to metric_binomial function, such as
citype
for type of confidence interval method.
Value
Depending on the detail
parameter, returns a numeric value
representing the calculated metric or a data frame/tibble with
detailed diagnostics including confidence intervals and possibly other
metrics relevant to understanding the metric.
Details
\(Accuracy = \frac{True Positives + True Negatives}{Total Cases}\)
Accuracy is one of the most intuitive performance measures and it is simply a ratio of correctly predicted observation to the total observations. It's a common starting point for evaluating the performance of a classifier. However, it's not suitable for unbalanced classes due to its tendency to be misleadingly high when the class of interest is underrepresented. For detailed diagnostics, including confidence intervals, specify detail = "full".
See also
dx_cm()
to understand how to create and interact with a
'dx_cm' object.
Examples
cm <- dx_cm(
dx_heart_failure$predicted,
dx_heart_failure$predicted,
threshold = 0.3, poslabel = 1
)
simple_accuracy <- dx_accuracy(cm, detail = "simple")
detailed_accuracy <- dx_accuracy(cm)
print(simple_accuracy)
#> [1] 0.532567
print(detailed_accuracy)
#> # A tibble: 1 × 8
#> measure summary estimate conf_low conf_high fraction conf_type notes
#> <chr> <chr> <dbl> <dbl> <dbl> <chr> <chr> <chr>
#> 1 Accuracy 53.3% (47.0%, 5… 0.533 0.470 0.594 139/261 Binomial… ""