MS-DS Master of Data science FREE Master of Data Science Machine Learning Questions and Answers 3

Question 1

Which dimensionality reduction technique preserves the global variance structure of the data by projecting onto orthogonal axes?

Accepted Answer

Principal Component Analysis (PCA)

Answer

PCA finds orthogonal principal components that maximize variance, making it ideal for linear dimensionality reduction while preserving global structure.

Question 2

What problem does the vanishing gradient issue cause during neural network training?

Accepted Answer

Earlier layers learn extremely slowly because gradients become negligibly small

Answer

Vanishing gradients cause the weights in earlier layers to receive near-zero updates, effectively preventing those layers from learning meaningful representations.

Question 3

Which evaluation metric is most appropriate when the cost of false negatives is significantly higher than false positives?

Accepted Answer

Recall

Answer

Recall measures the proportion of actual positives correctly identified, making it the priority metric when missing positive cases is costly.

Question 4

What is the key assumption of the Naive Bayes classifier?

Accepted Answer

All features are conditionally independent given the class label

Answer

Naive Bayes assumes that features are conditionally independent of each other given the class, which simplifies probability computation despite rarely being true in practice.

Question 5

In K-means clustering, how are cluster centroids updated during each iteration?

Accepted Answer

By computing the mean of all points currently assigned to each cluster

Answer

K-means updates each centroid by calculating the arithmetic mean of all data points assigned to that cluster in the current iteration.

MS-DS Master of Data science Practice Test

MS-DS Master of Data science Practice Test

MS-DS Master of Data science FREE Master of Data Science Machine Learning Questions and Answers 3