Question

Suppose that you have a binary classification problem, and your dataset has "n" attributes. How many parameters are there in the model?

a. n
b. n + 1
c. 2n
d. 2n + 1

Answer 1

In a binary classification model with 'n' attributes, there are 'n + 1' parameters, including one parameter per attribute and an extra bias term.

Step-by-step explanation:

In a binary classification problem with n attributes, the model will have n + 1 parameters. This count includes one parameter for each attribute and an additional parameter, often called the bias or intercept term. The correct answer is b. n + 1. This is because a binary classifier, such as logistic regression, calculates a weighted sum of the attribute values plus the bias, which is why we have n weights for the attributes and one extra parameter for the bias.