MCQs / Computer Science / NSCT: AI, Machine Learning & Data Analytics

NSCT: AI, Machine Learning & Data Analytics MCQs — Computer Science (page 4)

127 multiple-choice questions on NSCT: AI, Machine Learning & Data Analytics from Computer Science, each with the correct answer and, where available, a teacher's explanation. Read through them to revise the chapter, then practise the same questions one by one to test yourself.

76. A classifier's confusion matrix is: TP = 90, FP = 30, FN = 10, TN = 870. What is its precision?

  1. 25%
  2. 96%
  3. 75%
  4. 90%

Answer: C) 75%

Explanation: Precision = TP ÷ (TP + FP) = 90 ÷ 120 = 75%. Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

77. A classifier's confusion matrix is: TP = 60, FP = 15, FN = 20, TN = 105. What is its precision?

  1. 80%
  2. 82.5%
  3. 75%
  4. 20%

Answer: A) 80%

Explanation: Precision = TP ÷ (TP + FP) = 60 ÷ 75 = 80%. Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

78. A model scores 55% on both training and test data for a complex task. It is:

  1. Underfitting
  2. Leaking data
  3. Overfitting
  4. Ideal

Answer: A) Underfitting

Explanation: High bias – too simple. Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

79. A fraud model on data with 990 genuine and 10 fraud cases predicts "genuine" for every case. What is its accuracy and recall for fraud?

  1. Accuracy 1%, recall 100%
  2. Accuracy 99%, recall 0%
  3. Accuracy 50%, recall 50%
  4. Accuracy 99%, recall 99%

Answer: B) Accuracy 99%, recall 0%

Explanation: It is right on 990/1000 cases but catches none of the fraud – accuracy misleads on imbalanced data. Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

80. A value of 120 comes from data with mean 100 and standard deviation 20. What is its z-score?

  1. 20
  2. -1
  3. 2
  4. 1

Answer: D) 1

Explanation: z = (x − mean) ÷ std = (120 − 100) ÷ 20 = 1. Lesson 8.2 · AI, Machine Learning & Data Analytics

Discuss this question

81. Boosting methods like XGBoost build models:

  1. Independently in parallel only
  2. Without data
  3. By clustering
  4. Sequentially, each correcting the errors of the previous

Answer: D) Sequentially, each correcting the errors of the previous

Explanation: Bagging builds them in parallel. Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

82. A tree diagram showing how clusters merge in hierarchical clustering is a:

  1. ROC curve
  2. Dendrogram
  3. Confusion matrix
  4. Histogram

Answer: B) Dendrogram

Explanation: Cutting it at a height gives clusters. Lesson 8.4 · AI, Machine Learning & Data Analytics

Discuss this question

83. The Turing Test evaluates whether:

  1. A machine's conversation can be distinguished from a human's
  2. A program compiles
  3. A network is secure
  4. A database is normalized

Answer: A) A machine's conversation can be distinguished from a human's

Explanation: Proposed by Alan Turing in 1950. Lesson 8.1 · AI, Machine Learning & Data Analytics

Discuss this question

84. A classifier's confusion matrix is: TP = 60, FP = 15, FN = 20, TN = 105. What is its F1-score (to 2 decimal places)?

  1. 0.60
  2. 0.77
  3. 0.78
  4. 0.82

Answer: B) 0.77

Explanation: F1 = 2PR ÷ (P + R) = 2 × 0.80 × 0.75 ÷ (0.80 + 0.75) = 0.77. Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

85. A classifier's confusion matrix is: TP = 60, FP = 15, FN = 20, TN = 105. What is its accuracy?

  1. 82.5%
  2. 75%
  3. 80%
  4. 30%

Answer: A) 82.5%

Explanation: Accuracy = (TP + TN) ÷ total = 165 ÷ 200 = 82.5%. Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

86. The algorithm used to compute gradients of the loss with respect to all weights is:

  1. Apriori
  2. Minimax
  3. Backpropagation
  4. k-means

Answer: C) Backpropagation

Explanation: It applies the chain rule backwards. Lesson 8.5 · AI, Machine Learning & Data Analytics

Discuss this question

87. A plot that shows median, quartiles and outliers of a variable is a:

  1. Pie chart
  2. Scatter plot only
  3. Line chart
  4. Box plot

Answer: D) Box plot

Explanation: Points beyond 1.5 × IQR are outliers. Lesson 8.2 · AI, Machine Learning & Data Analytics

Discuss this question

88. A dataset where 98% of transactions are genuine and 2% fraudulent is:

  1. Perfectly balanced
  2. Unsupervised
  3. Imbalanced
  4. Normalized

Answer: C) Imbalanced

Explanation: Accuracy alone is misleading here. Lesson 8.2 · AI, Machine Learning & Data Analytics

Discuss this question

89. A hiring model trained on historical data rejects more women because past hiring was biased. This is a problem of:

  1. High recall
  2. Underfitting
  3. Data compression
  4. Algorithmic bias

Answer: D) Algorithmic bias

Explanation: Biased data produces biased decisions. Lesson 8.5 · AI, Machine Learning & Data Analytics

Discuss this question

90. Which Python library provides DataFrames for working with tabular data?

  1. Flask
  2. Requests
  3. Pandas
  4. NumPy only

Answer: C) Pandas

Explanation: Pandas is the core data-analysis library. Lesson 8.1 · AI, Machine Learning & Data Analytics

Discuss this question

91. Transformers, used in GPT and BERT, are based on:

  1. Decision trees
  2. Convolution only
  3. Self-attention
  4. k-means

Answer: C) Self-attention

Explanation: Attention relates every word to every other word. Lesson 8.5 · AI, Machine Learning & Data Analytics

Discuss this question

92. A model scores 99% on training data but 60% on test data. It is:

  1. Overfitting
  2. Not trained
  3. Underfitting
  4. Perfectly fitted

Answer: A) Overfitting

Explanation: It memorised the training data. Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

93. In k-NN with k = 7, the classes of the 7 nearest neighbours are: Cat, Dog, Dog, Dog, Cat, Cat, Dog. What class is predicted?

  1. Cat
  2. Dog
  3. A tie – no prediction
  4. The class of the farthest neighbour

Answer: B) Dog

Explanation: Majority vote: Dog (4 of 7). Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

94. A classifier's confusion matrix is: TP = 45, FP = 5, FN = 5, TN = 45. What is its precision?

  1. 89%
  2. 90%
  3. 91%
  4. 10%

Answer: B) 90%

Explanation: Precision = TP ÷ (TP + FP) = 45 ÷ 50 = 90%. Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

95. A classifier's confusion matrix is: TP = 8, FP = 2, FN = 12, TN = 978. What is its precision?

  1. 98.6%
  2. 80%
  3. 40%
  4. 20%

Answer: B) 80%

Explanation: Precision = TP ÷ (TP + FP) = 8 ÷ 10 = 80%. Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

96. A classifier's confusion matrix is: TP = 40, FP = 10, FN = 5, TN = 45. What is its accuracy?

  1. 88.9%
  2. 85%
  3. 80%
  4. 40%

Answer: B) 85%

Explanation: Accuracy = (TP + TN) ÷ total = 85 ÷ 100 = 85%. Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

97. A learning approach where an agent learns by receiving rewards and penalties is:

  1. Semi-supervised clustering
  2. Reinforcement learning
  3. Unsupervised learning
  4. Supervised learning

Answer: B) Reinforcement learning

Explanation: e.g. game-playing agents, robotics. Lesson 8.4 · AI, Machine Learning & Data Analytics

Discuss this question

98. Filling missing values with the column mean is called:

  1. Normalization
  2. Encoding
  3. Imputation
  4. Augmentation

Answer: C) Imputation

Explanation: Imputation replaces missing data. Lesson 8.2 · AI, Machine Learning & Data Analytics

Discuss this question

99. PCA reduces 784 features to 50 principal components. How many features does each sample have after PCA?

  1. 734
  2. 15
  3. 784
  4. 50

Answer: D) 50

Explanation: Each sample is projected onto the chosen components. Lesson 8.2 · AI, Machine Learning & Data Analytics

Discuss this question

100. A classifier's confusion matrix is: TP = 30, FP = 5, FN = 15, TN = 50. What is its recall?

  1. 85.7%
  2. 80%
  3. 33.3%
  4. 66.7%

Answer: D) 66.7%

Explanation: Recall = TP ÷ (TP + FN) = 30 ÷ 45 = 66.7%. Lesson 8.3 · AI, Machine Learning & Data Analytics

Discuss this question

Previous Page 4 of 6 Next

Practise NSCT: AI, Machine Learning & Data Analytics MCQs

More Computer Science chapters