24. 隐马尔可夫模型¶
24.1. 隐马尔可夫模型¶
24.1.1. 马尔可夫模型和朴素贝叶斯模型的关系¶
(24.1.1)¶\[p(y \mid \vec{x}) = \frac{p(y) \cdot p(\vec{x} \mid y)}{p(\vec{x})}\]
“the probability of a particular state is dependent only on the previous state”
\(p(y)\) 变成了 \(p(y_i|y_{i-1})\)
(24.1.2)¶\[p(\vec{y} \mid \vec{x}) = \prod_{i=1}^{n} p(y_{i} \mid y_{i-1}) \cdot p(x_{i} \mid y_{i})\]