K-Nearest Neighbors Quiz

What is the main idea behind K-Nearest Neighbors (KNN)?

• It builds a decision tree to classify data

• It uses similar data points to classify a new data point

• It minimizes a loss function during training

• It learns abstract parameters to generalize data

How does KNN make predictions?

• By using a pre-built model

• By minimizing an error function

• By comparing a new data point to its K closest neighbors

• By computing a weighted sum of all data points

Why is KNN considered a "lazy learner"?

• It trains slower than other algorithms

• It does not process data during prediction

• It stores training data and delays computation until prediction time

• It only works with small datasets

What does K in KNN represent?

• The total number of data points

• The number of neighbors considered for prediction

• The size of the training set

• The dimensionality of the dataset

What kind of distance is this?

• Manhattan

• Euclidean

What kind of distance is this?

• Manhattan

• Euclidean

What is an advantage of KNN?

• It is computationally efficient at prediction time

• It makes strong assumptions about the data distribution

• It can handle both classification and regression tasks

• It is robust to noisy data and outliers

What is a limitation of KNN?

• It requires preprocessing the data into binary form

• It is sensitive to noisy data and outliers

• It cannot handle regression problems

• It requires a large K value for every dataset

What happens if K is too large in KNN?

• The algorithm becomes highly sensitive to outliers

• The predictions become overly smooth and less accurate

• The algorithm fails to find neighbors for prediction

• It overfits the training data

Which of the following is NOT true about KNN?

• It is a supervised learning algorithm

• It is non-parametric and does not make assumptions about the data distribution

• It requires a distance metric to classify new data points

• It works by minimizing a loss function during training