Machine Learning Multiple Choice Questions and Answers for Practice

16. This clustering algorithm initially assumes that each data instance represents a single cluster.

1. conceptual clustering
2. K-Means clustering
3. expectation maximization
4. agglomerative clustering

17. This clustering algorithm merges and splits nodes to help modify nonoptimal partitions.

1. expectation maximization
2. conceptual clustering
3. K-Means clustering
4. agglomerative clustering

18. This supervised learning technique can process both numeric and categorical input attributes.

1. Bayes classifier
2. logistic regression
3. backpropagation learning
4. linear regression

19. This technique associates a conditional probability value with each data instance.

1. logistic regression
2. simple regression
3. multiple linear regression
4. linear regression

20. This unsupervised clustering algorithm terminates when mean values computed for the current iteration of the algorithm are identical to the computed mean values for the previous iteration.

1. conceptual clustering
2. K-Means clustering
3. expectation maximization
4. agglomerative clustering

21. When doing least-squares regression with regularisation (assuming that the optimisation can be done exactly), increasing the value of the regularisation parameter (Lambda)

1. will never increase the training error.
2. will never decrease the testing error.
3. will never increase the testing error.
4. will never decrease the training error.

22. Which is not true about Gradient of a continuous and differentiable function

1. is non-zero at a maximum
2. is zero at a saddle point
3. decreases as you get closer to the minimum
4. is zero at a minimum

23. Which of the following is a common use of unsupervised clustering?

1. determine a best set of input attributes for supervised learning
2. evaluate the likely performance of a supervised learner model
3. determine if meaningful relationships can be found in a dataset
4. detect outliers

24. Which of the following is not an advantage of Grid search

1. It can be applied to non-continuous functions.
2. It is easy to implement.
3. It runs reasonably fast for multiple linear regression.
4. It can be applied to non-differentiable functions.

25. Which of the following points would Bayesians and frequentists disagree on?

1. The use of probabilistic modelling for regression.
2. The use of prior distributions on the parameters in a probabilistic model.
3. The use of class priors in Gaussian Discriminant Analysis
4. The use of a non-Gaussian noise model in probabilistic regression.

26. Which of the following sentence is FALSE regarding regression?

1. It is used for prediction.
2. It may be used for interpretation.
3. It discovers causal relationships.
4. It relates inputs to outputs.

27. Which statement about outliers is true?

1. Outliers should be part of the training dataset but should not be present in the test data.
2. Outliers should be part of the test dataset but should not be present in the training data.
3. The nature of the problem determines how outliers are used.
4. Outliers should be identified and removed from a dataset.

28. Which statement is true about neural network and linear regression models?

1. Both models require numeric attributes to range between 0 and 1.
2. The output of both models is a categorical attribute value.
3. Both techniques build models whose output is determined by a linear sum of weighted input attribute values.
4. Both models require input attributes to be numeric.

29. Which statement is true about prediction problems?

1. The output attribute must be numeric.
2. The resultant model is designed to determine future outcomes.
3. The resultant model is designed to classify current behavior.
4. The output attribute must be categorical.

30. With Bayes classifier, missing data items are

1. treated as unequal compares.
2. replaced with a default value.
3. ignored.
4. treated as equal compares.