next up previous contents
Next: SQL Query Up: Integration of Data Mining Previous: Further Work

Proof of Bayes Rule

  Here we prove Bayes Rule from probability calculus. P(A) denotes the probability that event A will occur. Belief measure obeys the three basic axioms of probability theory:

\begin{displaymath}
0 \leq P(A) \leq 1
\end{displaymath}

\begin{displaymath}
P(sure\;proposition) = 1
\end{displaymath}

\begin{displaymath}
P(A\;or\;B) = P(A) + P(B)
\end{displaymath}

The proof derives from the multiplication rule of probability theory which is:

\begin{displaymath}
P(AB) = P(A \mid B)P(B)\end{displaymath}

Then:

\begin{displaymath}
P(H \mid e) = \frac{P(He)}{P(e)}
\end{displaymath}

By swapping He and using the multiplication rule again:

\begin{displaymath}
= \frac{P(e \mid H) P(H)}{P(e)}
\end{displaymath}

qed.


Torgeir Dingsoyr
2/26/1998