Virtual Labs

a: The number of features in the dataset. b: The distance metric used to calculate similarities. c: The number of nearest neighbors to consider for classification. d: The number of training instances in the dataset.

In the KNN algorithm, how is the class of a new instance determined?

a: By taking the weighted average of the classes of its k nearest neighbors. b: By selecting the class of the nearest neighbor with the highest confidence score. c: By choosing the class of the majority of its k nearest neighbors. d: By fitting a decision boundary using logistic regression.

Which of the following is a disadvantage of the KNN algorithm?

a: It is not suitable for multi-class classification. b: It requires a significant amount of memory to store the entire training dataset. c: It assumes that the data instances are linearly separable. d: It is sensitive to the feature scaling of the input data.

Which of the following statements about the decision boundary in K-Nearest Neighbors (KNN) algorithm is true?

a: The decision boundary in KNN is always linear. b: The decision boundary in KNN can be non-linear. c: The decision boundary in KNN depends only on the value of K. d: The decision boundary in KNN is determined by the feature space dimensionality.

Community Links Sakshat Portal Outreach Portal FAQ: Virtual Labs

AGPL 3.0 & Creative Commons (CC BY-NC-SA 4.0)