Unit 4: Regression and Clustering - Practice Quiz

INT394 — Machine Learning Algorithms 50 Questions
0 Correct 0 Wrong 50 Left
0/50

1 What is the fundamental difference between the target variables in classification and regression problems?

A. Classification predicts continuous values, while regression predicts discrete categories.
B. Both predict continuous values, but regression uses a different loss function.
C. Classification requires unsupervised learning, while regression requires supervised learning.
D. Classification predicts discrete class labels, while regression predicts continuous numerical values.

2 Which of the following scenarios is a regression problem?

A. Grouping customers into segments based on purchasing behavior.
B. Recognizing handwritten digits (0-9).
C. Predicting the price of a house based on its square footage.
D. Predicting whether an email is spam or ham.

3 In Simple Linear Regression, the relationship between the independent variable and the dependent variable is modeled as:

A.
B.
C.
D.

4 Which statement regarding Polynomial Regression is true?

A. It is a form of linear regression because it is linear in the parameters (coefficients).
B. It cannot be solved using Ordinary Least Squares (OLS).
C. It is considered a non-linear regression because the curve is non-linear.
D. It strictly requires non-parametric methods.

5 What happens if the degree of the polynomial in polynomial regression is chosen to be too high?

A. The model will overfit the data (High Variance).
B. The computational cost decreases significantly.
C. The model will underfit the data (High Bias).
D. The model will generalize better to unseen data.

6 Which loss function is most commonly used for Ordinary Least Squares (OLS) regression?

A. Kullback-Leibler Divergence
B. Mean Squared Error (MSE)
C. Cross-Entropy Loss
D. Hinge Loss

7 The Mean Squared Error (MSE) is calculated as:

A.
B.
C.
D.

8 Which loss function is more robust to outliers in a regression problem?

A. Mean Absolute Error (MAE)
B. Root Mean Squared Error (RMSE)
C. L2 Norm
D. Mean Squared Error (MSE)

9 In the context of regression regularization, Lasso Regression adds which penalty term to the loss function?

A. L2 penalty (Squared magnitude of coefficients: )
B. L1 penalty (Absolute magnitude of coefficients: )
C. A combination of L1 and L2 penalties
D. No penalty term

10 What is a defining characteristic of Non-Parametric Regression?

A. It assumes a fixed mathematical form (e.g., a line) with a finite set of parameters.
B. The number of parameters grows with the size of the training data.
C. It requires the data to be normally distributed.
D. It only works for classification problems.

11 In K-Nearest Neighbors (KNN) regression, how is the prediction for a new data point made?

A. By taking the majority vote of the class labels of neighbors.
B. By solving a linear equation .
C. By taking the average (or weighted average) of the target values of the 'K' closest training neighbors.
D. By calculating the probability using Bayes' theorem.

12 Which of the following is true regarding the choice of 'k' in KNN regression?

A. A very small 'k' (e.g., k=1) leads to high variance (overfitting).
B. A very large 'k' leads to overfitting (high variance).
C. The value of 'k' does not affect the model performance.
D. A very small 'k' (e.g., k=1) leads to high bias (underfitting).

13 What is the primary difference between Supervised Learning (Classification/Regression) and Unsupervised Learning (Clustering)?

A. Supervised learning requires labeled data (input-output pairs), while unsupervised learning uses unlabeled data.
B. Supervised learning is faster than unsupervised learning.
C. Unsupervised learning always yields better accuracy.
D. Supervised learning groups data, while unsupervised learning predicts values.

14 The Euclidean distance between two points and is given by:

A.
B.
C.
D.

15 Which distance measure corresponds to the norm and is calculated as the sum of absolute differences?

A. Chebyshev Distance
B. Cosine Distance
C. Manhattan Distance
D. Euclidean Distance

16 Cosine Similarity is particularly useful for:

A. Calculating distance on a grid.
B. Time series forecasting.
C. Geometric clustering of low-dimensional data.
D. Measuring the similarity between text documents (represented as vectors) irrespective of magnitude.

17 The Minkowski distance is a generalization of both Euclidean and Manhattan distances defined as . If , it becomes:

A. Chebyshev Distance
B. Mahalanobis Distance
C. Euclidean Distance
D. Manhattan Distance

18 Which of the following is a Partition-based clustering algorithm?

A. K-Means
B. DBSCAN
C. Agglomerative Clustering
D. BIRCH

19 What is the objective function that the K-Means algorithm tries to minimize?

A. The number of clusters
B. Silhouette Coefficient
C. Within-Cluster Sum of Squares (WCSS)
D. Between-Cluster Sum of Squares

20 Which of the following is a step in the K-Means algorithm?

A. Selecting the 'k' nearest neighbors for voting.
B. Assigning points to the nearest cluster centroid.
C. Drawing a separating hyperplane.
D. Merging the two closest clusters.

21 A major limitation of the standard K-Means algorithm is:

A. It is computationally very expensive for small datasets.
B. It requires the number of clusters to be specified in advance.
C. It always finds the global optimum.
D. It works well with non-convex cluster shapes.

22 How does K-Medoids differ from K-Means?

A. K-Medoids uses the mean of the points as the center.
B. K-Medoids uses Euclidean distance exclusively.
C. K-Medoids uses actual data points as centers (medoids) and is more robust to outliers.
D. K-Medoids is faster than K-Means.

23 Hierarchical clustering can be divided into two main types:

A. Agglomerative (Bottom-Up) and Divisive (Top-Down)
B. Linear and Non-linear
C. Centroid-based and Density-based
D. Supervised and Unsupervised

24 In Agglomerative Hierarchical Clustering, what does 'Single Linkage' measure?

A. The minimum distance between the closest pair of points in two clusters.
B. The distance between the centroids of two clusters.
C. The maximum distance between points in two clusters.
D. The average distance between all pairs of points in two clusters.

25 What is a Dendrogram?

A. A plot showing the loss function over iterations.
B. A method to calculate the derivative of a function.
C. A diagram representing the tree structure of hierarchical clustering.
D. A scatter plot of the clusters.

26 In hierarchical clustering, 'Complete Linkage' uses which distance metric to merge clusters?

A. Distance between centroids.
B. Average distance between points.
C. Minimum distance between points (nearest neighbors).
D. Maximum distance between points (farthest neighbors).

27 Which clustering method does NOT require specifying the number of clusters upfront?

A. Hierarchical Clustering
B. K-Means
C. Gaussian Mixture Models
D. K-Medoids

28 What is the Elbow Method used for?

A. To prevent overfitting in regression.
B. To determine the optimal number of clusters () in K-Means.
C. To calculate the distance between clusters.
D. To visualize high-dimensional data.

29 The Silhouette Score ranges between:

A. -infinity and +infinity
B. -1 and 1
C. 0 and 1
D. 0 and 100

30 A Silhouette Score close to +1 implies:

A. The point is on or very close to the decision boundary between two neighboring clusters.
B. The clustering algorithm failed.
C. The point is well matched to its own cluster and far from neighboring clusters.
D. The point is assigned to the wrong cluster.

31 Which metric is used for cluster validation when ground truth labels are available?

A. Rand Index (or Adjusted Rand Index)
B. Elbow Method
C. Davies-Bouldin Index
D. Silhouette Score

32 In the context of Ridge Regression, as the penalty parameter approaches infinity, the regression coefficients tend towards:

A. Infinity
B. The OLS estimates
C. Zero
D. 1

33 Which regression technique fits a local regression model to a subset of the data surrounding the query point?

A. LOESS (Locally Estimated Scatterplot Smoothing)
B. Ridge Regression
C. Logistic Regression
D. Linear Regression

34 Jaccard Similarity is defined as:

A.
B.
C.
D.

35 K-Means++ is an algorithm used for:

A. Initializing the cluster centers to improve convergence speed and quality.
B. Calculating the final centroids.
C. Determining the value of K automatically.
D. Post-processing the clusters.

36 Which of the following data shapes is K-Means least likely to handle correctly?

A. Spherical clusters of equal size.
B. Clusters with similar variances.
C. Concentric circles (e.g., a donut shape).
D. Compact, well-separated blobs.

37 The Dunn Index is an internal cluster validation metric where a higher value indicates:

A. Loose and overlapping clusters.
B. High computational complexity.
C. Poor clustering performance.
D. Compact and well-separated clusters.

38 Which statement regarding the bias-variance trade-off in regression is correct?

A. We want to maximize both bias and variance.
B. Complex non-linear models usually have low bias and high variance.
C. Variance refers to the error on the training set.
D. Simple linear models usually have low bias and high variance.

39 What is Ward's Method in hierarchical clustering?

A. A divisive method that splits based on density.
B. A method that uses random linkage.
C. An agglomerative method that minimizes the increase in total within-cluster variance when merging.
D. A method equivalent to single linkage.

40 Hamming distance is primarily used for:

A. Image pixel intensity.
B. Categorical data or strings of equal length.
C. Geospatial coordinates.
D. Continuous numerical data.

41 In kernel regression (e.g., Nadaraya-Watson), the 'bandwidth' parameter controls:

A. The smoothness of the fit (width of the kernel window).
B. The number of iterations.
C. The learning rate of the gradient descent.
D. The number of clusters.

42 Which of the following is NOT a metric for calculating the distance between two clusters in hierarchical clustering?

A. Single Linkage
B. Average Linkage
C. Complete Linkage
D. Gradient Descent

43 For a dataset with points, what is the time complexity of one iteration of K-Means with clusters and dimensions?

A.
B.
C.
D.

44 What is the main advantage of Hierarchical Clustering over K-Means?

A. It handles missing values natively.
B. It scales linearly with the number of data points.
C. It provides a taxonomy/hierarchy of clusters and doesn't require pre-specifying .
D. It is computationally faster for large datasets.

45 If a regression model has an (Coefficient of Determination) score of 1.0, it means:

A. The model perfectly fits the data.
B. The model explains none of the variability of the response data.
C. The model is a constant line.
D. The model is underfitting.

46 Which of these is a 'lazy learning' algorithm often used for regression?

A. Linear Regression
B. K-Nearest Neighbors (KNN)
C. K-Means
D. Ridge Regression

47 In the context of clustering, what is 'inter-cluster distance'?

A. The sum of squared errors.
B. The distance between points within the same cluster.
C. The distance between different clusters.
D. The distance from a point to the origin.

48 When using the Manhattan distance, the set of points at a constant distance from the origin forms a:

A. Square (rotated 45 degrees)
B. Sphere
C. Hyperbola
D. Circle

49 Which statement regarding outlier sensitivity is correct?

A. K-Means is less sensitive to outliers than K-Medoids.
B. Least Squares Regression is robust to outliers.
C. Median-based methods are more sensitive to outliers than Mean-based methods.
D. K-Means is sensitive to outliers because the mean is influenced by extreme values.

50 What is the 'Kernel Trick' in the context of non-linear regression (e.g., Support Vector Regression)?

A. Mapping data to a higher-dimensional space to make it linearly separable/fittable without explicitly calculating coordinates.
B. Using a GPU kernel for faster processing.
C. A method to reduce dimensionality.
D. Ignoring non-linear data points.