1
کامپیوتر و شبکه::
نرخ عدم پذیرش اشتباه
The false reject rate (FRR), also known as Type I error [25], measures the probability of an enrolled individual not being identified by the system.
The false accept and false reject rates can be calculated by the amount of overlap between two distributions, which is illustrated in Figure 5.3.
FRR = j (5.4)
The false accept rate is defined by the normalised area between 0 and the separation point, K, in the inter-class distribution Pdiff- The false reject rate is defined as the normalised area between the separation point, K, and 1 in the intra-class distribution Psarne.
Clearly the separation point will influence the false accept and false reject rates, since a lower separation Hamming distance will decrease FAR while increasing FRR, and vice versa.
واژگان شبکه مترجمین ایران