Short Explanation
P stand for Positives (Red) and N stand for Negatives (Blue).
TP: True Positive, FP: False positive, FN: False Negative, TN: True Negative.
The goal of this program is to separate the data into its correct class (red and blue) by changing the threshold.
On the right canvas different metrics are shown to evaluate how good the classification can be.
ROC
The receiver operating characteristic (ROC) curve plots the true positive rate (TPR) against the false positive rate (FPR). If the curve is very flat, then there is no good way to classify the data.
Loss Functions
The 0-1-Loss counts the number of points which are wrongly classified.
The Logistical Loss uses following formula:
log(1+exp(-yf(x)))
where f(x) is the distance to the threshold and y is either +1 or -1 (depending on the class).
source: wiki