The 120 values in the lower-left corner indicate mushrooms that are actually edible but were classified as poisonous.
- Example – identifying poisonous mushrooms with rule learners
- from Machine Learning with R
- Publisher: Packt Publishing
- Released: October 2013
This is incorrect. The 120 values in the lower left were classified as edible when they were actually poisonous. To validate, run table(mushrooms$type), and you will see that the original data listed 4,208 as edible, and 3,916 as poisonous. In the confusion matrix, you see the 4,208 that were listed as edible on line 1, and the 120 + 3,796 = 3,916 are listed as poisonous on line 2, which corresponds to the actual values. Since we know the 3,916 are actually poisonous, and 120 mushrooms were classified incorrectly per summary(mushroom_1R), we can see that the 120 value in the lower left of the confusion matrix must have been classified as edible when they were actually poisonous.
Share this highlighthttp://www.safaribooksonline.com/a/machine-learning-with/68455/