Why sometimes accuracy is not a good measure to check efficiency of machine learning algorithms?

Why sometimes accuracy is not a good measure to check efficiency of machine learning algorithms?

Accuracy can be a useful measure if we have the same amount of samples per class but if we have an imbalanced set of samples accuracy isn’t useful at all. Even more so, a test can have a high accuracy but actually perform worse than a test with a lower accuracy.

How do you measure the accuracy of a classifier?

You simply measure the number of correct decisions your classifier makes, divide by the total number of test examples, and the result is the accuracy of your classifier.

Why is predictive accuracy a misleading performance measure?

However, as illustrated above, predictive accuracy can be a misleading in the presence of class-imbalance. In such cases, more weights are placed on the majority class than on the minority class making it more difficult for a classifier to perform well on the minority class.

Why is accuracy not the best measure of accuracy?

Accuracy, the proportion of correct classifications among all classifications, is very simple and very “intuitive” measure, yet it may be a poor measure for imbalanced data. Why does our intuition misguide us here and are there any other problems with this measure? machine-learningmodel-evaluationaccuracyscoring-rulesfaq

How is the accuracy of a prediction maximized?

Accuracy is maximized if we classify everything as the first class and completely ignore the 40% probability that any outcome might be in the second class. (Here we see that accuracy is problematic even for balanced classes.) Proper scoring-rules will prefer a (0.6, 0.4) prediction to the (1, 0) one in expectation.

Is the accuracy of a classifier a misleading measure?

Most of the performance metrics utilized in classification problems are derived based on the confusion matrix. The most commonly reported measure of a classifier is the accuracy. This measure evaluates the overall efficiency of an algorithm. However, as illustrated above, predictive accuracy can be a misleading in the presence of class-imbalance.