巨硬 data science intern挂机

collwe 发表于 2019-1-3 07:50
我没碰到，语言随便你选

多谢楼主！
你说的那个贝叶斯题目 我在glassdoor上面找到了 说的感觉跟你的很像， 因为他这个也是少个概率所以他就用p表示了， 不知道你最后也是这么算的吗？

With Bayes' rule,

P(r | y, y, y) = P(y,y,y | r) * P(r) / (P(y,y,y | r)*P(r) + P(y,y,y | not r)*(1-P(r))

Let's write P(r) = p (probability that it is raining)
Moreover, we have P(y,y,y|r) = P(y|r)^3 = (2/3)^3 (Truth hypothesis + Bayes and independence)
and P(y,y,y|not r) = (1/3)^3 (same)

Then, plugging it into the bayes' formula, we have

P(r | y,y,y) = 8 p / (7p + 1)

Note that if p = 0 (say Seattle is in the middle of the Sahara, the probability of raining is always zero, independently from the three answers)

麻烦楼主了！