Working with Probabilities Flashcards
Conditional Dependency
Probability of A given B
P ( A | B )
P(A and B) / P(B)
Independent Events
Probability of A and B
P(A) * P(B)
Independent Events
Probability of A or B
P(A) + P(B)
Bayes’ Rules
Use to flip conditional probabilities
P( A | B ) = P( B | A ) * P(A) / P(B)
Maximum Likelihood Estimation
Estimating probability from relative frequency
Probability of A given B
= count ( A & B ) / count (B)
The Chain Rule
P (A1 & A2 & … & An-1 & An)
= P(An | An-1,…,A1) * P (An-1, …, A1)
P (A,B,C,D)
= P( D | C,B,A ) * P( C | B,A ) * P( B | A ) * P(A)
The Markov Assumption
When we’re looking at the probability of An given A1 through An-1, we only need to look a few words back
You only need to look at the previous few words to estimate the probability of the current word
P (An | An-1, An-2)