I have this data frame called conf_mat with two columns including predicted values and reference values in each objects. I have 20 objects in this dataframe.
dput(Conf_mat)
structure(list(Predicted = c(100, 200, 200, 100, 100, 200, 200,
200, 100, 200, 500, 100, 100, 100, 100, 100, 100, 100, 500, 200
), Reference = c(600, 200, 200, 200, 200, 200, 200, 200, 500,
500, 500, 200, 200, 200, 200, 200, 200, 200, 200, 200)), .Names = c("Predicted",
"Reference"), row.names = c(NA, 20L), class = "data.frame")
I want to create a confusion matrix out of this table with this kind of structure which will be filled in by the Conf_mat dataframe. This will be allow me to compute an accuracu assessment of my classification. Thanks for your help.
100 200 300 400 500 600
100 NA NA NA NA NA NA
200 NA NA NA NA NA NA
300 NA NA NA NA NA NA
400 NA NA NA NA NA NA
500 NA NA NA NA NA NA
600 NA NA NA NA NA NA
1) Try the following:
table(Conf_mat)
2) If you want to force levels 100, 200, ..., 600 to appear:
conf_mat_tab <- table(lapply(Conf_mat, factor, levels = seq(100, 600, 100)))
3) You could also try this:
library(caret)
confusionMatrix(conf_mat_tab) # conf_mat_tab from (2)
which gives:
Confusion Matrix and Statistics
Reference
Predicted 100 200 300 400 500 600
100 0 9 0 0 1 1
200 0 6 0 0 1 0
300 0 0 0 0 0 0
400 0 0 0 0 0 0
500 0 1 0 0 1 0
600 0 0 0 0 0 0
Overall Statistics
Accuracy : 0.35
95% CI : (0.1539, 0.5922)
No Information Rate : 0.8
P-Value [Acc > NIR] : 1
Kappa : 0.078
Mcnemar's Test P-Value : NA
Statistics by Class:
Class: 100 Class: 200 Class: 300 Class: 400 Class: 500 Class: 600
Sensitivity NA 0.3750 NA NA 0.3333 0.00
Specificity 0.45 0.7500 1 1 0.9412 1.00
Pos Pred Value NA 0.8571 NA NA 0.5000 NaN
Neg Pred Value NA 0.2308 NA NA 0.8889 0.95
Prevalence 0.00 0.8000 0 0 0.1500 0.05
Detection Rate 0.00 0.3000 0 0 0.0500 0.00
Detection Prevalence 0.55 0.3500 0 0 0.1000 0.00
Balanced Accuracy NA 0.5625 NA NA 0.6373 0.50