The GATE Grind

GATE 2024 DA – Question 20

Machine Learning · Supervised learning: classification · 1 mark · Multiple choice

Given a dataset with $K$ binary-valued attributes (where $K > 2$) for a two-class classification task, the number of parameters to be estimated for learning a naïve Bayes classifier is

  1. $2K + 1$
  2. $2^K + 1$
  3. $2^{K+1} + 1$
  4. $K^2 + 1$

Practise this question in The GATE Grind →

Show answer and explanation

Correct answer: (A) $2K + 1$

Explanation

Naive Bayes needs $P(x_i = 1 | \text{class})$ for each of the $K$ attributes and each of the 2 classes, which is $2K$ parameters, and the prior $P(\text{class} = 1)$, one more. The total is $2K + 1$.