Experiment – 2
Name: Muskan Soni UID: 22BCS16851
Branch: BE-CSE Section/Group: DL-902/B
Semester: 6th Date:
Subject: Deep Learning Lab Subject Code: 22CSP-368
1. Aim: To implement Bayes' theorem using Python to compute posterior
probabilities and demonstrate its application in decision-making.
2. Objective:
• To understand and apply Bayes' theorem to infer probabilities based on prior
knowledge and observed data.
• To calculate the posterior probability for an event.
Bayes' theorem is mathematically expressed as:
[ P(H|E) = \frac{P(E|H) \cdot P(H)}{P(E)} ]
Where:
• ( P(H|E) ): Posterior Probability - The probability of the hypothesis ( H ) being true
given the evidence ( E ).
• ( P(E|H) ): Likelihood - The probability of observing the evidence ( E ) given that the
hypothesis ( H ) is true.
• ( P(H) ): Prior Probability - The initial probability of the hypothesis ( H ) before
observing the evidence.
• ( P(E) ): Marginal Likelihood - The total probability of observing the evidence ( E )
under all possible hypotheses.
3. Implementation:
def bayes_theorem(prior_h, likelihood_e_given_h, likelihood_e_given_not_h):
prior_not_h = 1 - prior_h
p_e = (likelihood_e_given_h * prior_h) + (likelihood_e_given_not_h * prior_not_h)
posterior_h_given_e = (likelihood_e_given_h * prior_h) / p_e
return posterior_h_given_e
# Given values
prior_h = 0.01 # P(H)
likelihood_e_given_h = 0.9 # P(E|H)
likelihood_e_given_not_h = 0.05 # P(E|¬H)
# Calculate posterior probability
posterior_probability = bayes_theorem(prior_h, likelihood_e_given_h,
likelihood_e_given_not_h)
print(f"The posterior probability of having the disease given a positive test result is:
{posterior_probability:.4f}")
def bayes_theorem(prior_h, likelihood_e_given_h, likelihood_e_given_not_h):
prior_not_h = 1 - prior_h
p_e = (likelihood_e_given_h * prior_h) + (likelihood_e_given_not_h * prior_not_h)
posterior_h_given_e = (likelihood_e_given_h * prior_h) / p_e
return posterior_h_given_e
# Given values for spam classification
prior_h = 0.2 # P(H): Probability that an email is spam
likelihood_e_given_h = 0.8 # P(E|H): Probability of "discount" given spam
likelihood_e_given_not_h = 0.1 # P(E|¬H): Probability of "discount" given not spam
# Calculate posterior probability
posterior_probability = bayes_theorem(prior_h, likelihood_e_given_h,
likelihood_e_given_not_h)
print(f"The posterior probability that the email is spam given it contains 'discount' is:
{posterior_probability:.4f}")
4. Output:
5. Learning Outcomes:
• Result: For this example, the posterior probability of having the disease given
a positive test result is approximately 0.17 (17%).
• Insight: Even with a highly sensitive and specific test, the low prevalence of
the disease significantly reduces the posterior probability.
