MLS-C01 Question #115: Real Exam Question with Answer & Explanation

Question

A Machine Learning Specialist is deciding between building a naive Bayesian model or a full Bayesian network for a classification problem. The Specialist computes the Pearson correlation coefficients between each feature and finds that their absolute values range between 0.1 to 0.95. Which model describes the underlying data in this situation?

Options

• AA naive Bayesian model, since the features are all conditionally independent.
• BA full Bayesian network, since the features are all conditionally independent.
• CA naive Bayesian model, since some of the features are statistically dependent.
• DA full Bayesian network, since some of the features are statistically dependent.

Explanation

{"question_number": 1, "correct_answer": "D", "explanation": "A naive Bayes classifier assumes that all features are conditionally independent of one another given the class label. Pearson correlation coefficients measure linear dependence between features. When absolute values range from 0.1 to 0.95, many pairs of features have strong correlations (up to 0.95), indicating they are statistically dependent - violating the naive Bayes independence assumption. A full Bayesian network, by contrast, explicitly models conditional dependencies between features using a directed acyclic graph (DAG), making it appropriate when features are not all independent. Options A and B are incorrect because the data clearly contains dependent features (high correlations). Option C is wrong because naive Bayes cannot properly model statistically dependent features. Option D is correct: the feature dependencies revealed by the correlation analysis require a full Bayesian network.", "generated_by": "claude-sonnet", "llm_judge_score": 4}

No community discussion yet for this question.