Machine Learning Algorithms MCQs

Machine Learning Algorithms MCQs with answers are exclusively curated for Class 11 Artificial Intelligence students. Every question is aligned with CBSE curriculum and covers all question types and patterns, so you can practise with confidence, sharpen your concepts, and score higher in exams.

Q1. An online shopping application recommends socks to a customer after they purchase shoes. Which technology enables the application to identify such purchasing patterns?
a) Traditional programming
b) Machine Learning
c) Manual data entry
d) Database management

Q2. What distinguishes Machine Learning from traditional programming?
a) ML requires developers to provide precise instructions for every decision.
b) ML learns patterns and relationships from data.
c) ML works only with numerical data.
d) ML cannot make predictions on unseen data.

Q3. A machine is trained using examples containing both input data and their correct outputs. Which learning approach is being used?
a) Unsupervised Learning
b) Reinforcement Learning
c) Supervised Learning
d) Clustering

Q4. A system examines an unlabelled dataset to discover hidden patterns and relationships. Which type of Machine Learning is appropriate?
a) Supervised Learning
b) Unsupervised Learning
c) Reinforcement Learning
d) Regression

Q5. A robot learns to perform tasks by interacting with its environment and receiving rewards or penalties. Which learning method does it use?
a) Classification
b) Supervised Learning
c) Reinforcement Learning
d) Unsupervised Learning

Q6. Which of the following is NOT a main type of machine learning?
a) Supervised learning
b) Unsupervised learning
c) Reinforcement learning
d) Deductive learning

Q7. A model performs very well on training data but poorly on new data. This problem is called:
a) Clustering
b) Overfitting
c) Regression
d) Reinforcement

Q8. A model whose working is difficult to interpret is said to act as a:
a) Glass box
b) Decision tree
c) Black box
d) Neighbour

Q9. Bias in training data mainly results in:
a) Faster training
b) Biased predictions
c) Perfect accuracy
d) Fewer features

Q10. In supervised learning, the input data is accompanied by:
a) Rewards and penalties
b) Random noise
c) The correct output (labels)
d) No additional information

Q11. The main goal of unsupervised learning is to:
a) Predict labels for new data
b) Find hidden patterns or structure in unlabelled data
c) Maximize cumulative rewards
d) Draw the regression line

Q12. Which of the following is an example of a reinforcement learning algorithm?
a) k-means
b) Linear regression
c) Q-learning
d) Hierarchical clustering

Q13. Which of the following is a supervised learning algorithm?
a) k-means clustering
b) Hierarchical clustering
c) Principal component analysis
d) Support vector machine

Q14. Regression is used when the output variable is:
a) Discrete
b) Continuous
c) Unlabelled
d) A reward

Q15. Classification works with:
a) Continuous data
b) Discrete data
c) Only images
d) Only text

Q16. A company wants to predict the monthly sales revenue based on advertising expenditure. Which type of supervised learning should it use?
a) Classification
b) Clustering
c) Regression
d) Reinforcement Learning

Q17. Correlation measures the:
a) Strength of a linear relationship between two quantitative variables
b) Number of clusters in a dataset
c) Number of outliers
d) Distance between two centroids

Q18. In a positive correlation:
a) Variables move in opposite directions
b) Variables move in the same direction
c) There is no relationship between the variables
d) One variable stays constant

Q19. If changes in one variable do not indicate any predictable changes in another variable, the relationship is called:
a) Positive correlation
b) Negative correlation
c) Zero correlation
d) Linear regression

Q20. A Pearson’s r of -0.9 indicates:
a) High negative correlation
b) Low positive correlation
c) No correlation
d) Perfect positive correlation

Q21. If Pearson’s r = 0, it means:
a) Perfect positive association
b) Perfect negative association
c) No association between the two variables
d) Data has no outliers

Q22. The value of Pearson’s r always lies between:
a) 0 and 1
b) -1 and +1
c) -10 and +10
d) 0 and 100

Q23. Which is NOT a requirement for using Pearson’s correlation coefficient?
a) Scale of measurement should be interval or ratio
b) Variables should be approximately normally distributed
c) The association should be linear
d) The data should contain several outliers

Q24. In simple regression, the variable y is also called the:
a) Independent variable
b) Predictor variable
c) Explanator variable
d) Dependent (explained) variable

Q25. In the equation y = a + bx + e, the term b represents the:
a) Intercept
b) Slope
c) Error
d) Mean

Q26. In y = a + bx + e, the term e represents:
a) The error or residual
b) The slope
c) The independent variable
d) The number of clusters

Q27. The least squares method finds the best-fit line by minimizing:
a) The number of data points
b) The sum of squared differences between observed and predicted values
c) The distance between centroids
d) The number of outliers

Q28. The regression line passes through the:
a) Origin only
b) Maximum value of y
c) Mean of the independent and dependent variables
d) Smallest data point

Q29. Multiple linear regression uses:
a) Only one independent variable
b) More than one independent variable
c) No independent variable
d) Only categorical variables

Q30. Which is an advantage of linear regression?
a) It handles every non-linear relationship
b) It is unaffected by outliers
c) It is a simple technique and easy to implement
d) It requires no data

Q31. Which is a disadvantage of linear regression?
a) It is limited to linear relationships between variables
b) It is difficult to implement
c) It cannot be used for prediction
d) It needs unlabelled data

Q32. Which of these is an application of linear regression?
a) Predicting salary based on years of experience
b) Grouping documents by content similarity
c) Training an agent to play a game
d) Building a dendrogram

Q33. For the Python program with x = [1, 2, 3, 4, 5] and y = [2, 4, 5, 4, 5], the expected slope is:
a) 1.75
b) 0.35
c) 0.75
d) 4.00

Q34. For the same program, the expected y-intercept is:
a) 0.75
b) 1.75
c) 3.00
d) 4.00

Q35. The main objective of classification is to:
a) Predict continuous values
b) Assign labels to data instances based on their features
c) Group unlabelled data without any labels
d) Reduce the number of variables

Q36. Sorting waste into paper, metal and plastic bins is an example of:
a) Regression
b) Clustering
c) Classification
d) Reinforcement learning

Q37. The dataset used to build a classification model, containing labelled examples, is called:
a) Test data
b) Training data
c) Noise data
d) Cluster data

Q38. Email spam detection (spam or not spam) is an example of:
a) Binary classification
b) Multi-class classification
c) Multi-label classification
d) Clustering

Q39. Plant species classification is an example of:
a) Binary classification
b) Multi-class classification
c) Multi-label classification
d) Imbalanced classification

Q40. Identifying all objects (bicycle, apple, person) present in one photo is an example of:
a) Binary classification
b) Multi-class classification
c) Multi-label classification
d) Regression

Q41. Fraud detection is a typical example of:
a) Binary classification
b) Multi-label classification
c) Imbalanced classification
d) Clustering

Q42. k-NN is described as a:
a) Non-parametric supervised learning technique
b) Reinforcement learning technique
c) Unsupervised clustering technique
d) Dimensionality reduction technique

Q43. In k-NN, the distance calculated between data points is the:
a) Manhattan distance
b) Euclidean distance
c) Time difference
d) Cosine angle

Q44. In k-NN, the new data point is assigned to the category that:
a) Has the fewest neighbours among the K nearest
b) Has the most neighbours among the K nearest
c) Is the farthest away
d) Is chosen randomly

Q45. Which is an advantage of k-NN?
a) It has no explicit training phase
b) It is cheap for very large datasets
c) It needs no feature scaling
d) It works best on high-dimensional data

Q46. Which is a limitation of k-NN?
a) It cannot be used for classification
b) It is computationally expensive, especially for large datasets
c) It is easy to understand
d) It is robust to outliers

Q47. Clustering is a technique used to group:
a) Labelled datasets
b) Unlabelled datasets based on similarity
c) Only numerical data
d) Only text data

Q48. The first step in the clustering process is to:
a) Interpret the results
b) Run the clustering algorithm
c) Prepare the data
d) Cut the dendrogram

Q49. Partitioning clustering is also known as the:
a) Density-based method
b) Centroid-based method
c) Tree-based method
d) Probability-based method

Q50. Density-based clustering faces difficulty when the dataset has:
a) Varying densities and high dimensions
b) Very few points
c) Only two features
d) Labelled points

Q51. A dendrogram is associated with:
a) k-NN
b) Linear regression
c) Hierarchical clustering
d) Q-learning

Q52. Which is the most common example of hierarchical clustering?
a) Agglomerative Hierarchical algorithm
b) k-means algorithm
c) k-NN algorithm
d) Linear regression

Q53. In k-means clustering, K stands for:
a) The number of features
b) The number of clusters
c) The number of outliers
d) The number of iterations always required

Q54. Grouping the pixels of an image into regions of similar colours is an application of k-means known as:
a) Anomaly detection
b) Image segmentation
c) Sales forecasting
d) Fraud detection