224 In 2014, Facebook researchers published a paper describing a facial recognit

Question

224 In 2014, Facebook researchers published a paper describing a facial recognition that achieves a success rate of about 97.3% correct. i a Normally a test like this is described by two numbers, typically the false alarm bility and the miss probability. The usual way of combining these two numbers into a single number is q = max (Prl False Alarm], Pr(Miss), where q = 1-2 = 2.7% Facebook system. In this case, what can be said about the false alarm and mis ities? In addition to their numerical values, describe in words what they mean. for the probabil b. The Facebook result assumes high-quality images. In practical situations, the success is likely to be much lower. Nevertheless, assume a facial recognition system like this is used at a sporting event with n = 70,000 spectators to identify a possible "bad guy" who may or may not be at the event. Assuming the false alarm and miss rates are both equal to q = 2.7%, describe the result that will likely occur (i) if the bad guys present and (i) if the bad guy is not present.

Explanation / Answer

a) False alarm means incorrectly selecting a wrong person

Miss means incorrectly rejecting a right person

b) Number of people identified as bad/ Bad guy presnet) = 69999*2.7%+97.3%= 1890.946

Number of people identified as bad/ bad guy absent = 70000*2.7%=1890

In other words, there is not much difference in identifying good guys as bad guys, even if bad guy is present or not

Related Questions

Navigate

• Browse (All)
• Browse 2
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.