Scribd
Upload
9 views

Machine Learning

The document outlines the instructions and structure for an examination in Machine Learning at ICOF Colleges. It consists of three sections: multiple-choice questions, short answer questions, and essay questions, covering various topics in machine learning. Candidates are required to answer all questions in sections A and B, and choose one question from section C.

Uploaded by

Sub Phiri
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
9 views

Machine Learning

The document outlines the instructions and structure for an examination in Machine Learning at ICOF Colleges. It consists of three sections: multiple-choice questions, short answer questions, and essay questions, covering various topics in machine learning. Candidates are required to answer all questions in sections A and B, and choose one question from section C.

Uploaded by

Sub Phiri
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

NAME: ………………………… ID NUMBER:………………

ICOF Colleges Seminary and Universities

COMPUTER SCIENCE
MACHINE LEARNING
AUGUST EXAMINATIONS
_______________________________________________________________________
TIME: 1HOURS 30 MINUTES MARKS: 100

INSTRUCTIONS TO CANDIDATES
1. Write your name and your Student’s ID number in the space at the top of this page.
2. This paper consists of three (3) sections, A, B and C.
3. Answer all questions in section A and B.
4. Circle the letters representing your choices for section A questions.
5. Write the correct answers for Section B questions in the spaces provided.
6. Section C has two (2) questions. Answer any one question of your choice.
7. Cell phones and programmable devices should not be brought into the examination room.

FOR EXAMINER’S USE MARKS


A

TOTAL

Page 1 of 5
DO NOT TURN THIS PAGE UNTIL YOU ARE TOLD TO DO SO.
SECTION A (10 MARKS)

1. Which of the following is not a type of machine learning? A) Supervised learning B)


Unsupervised learning C) Reinforcement learning D) Structured learning
2. In supervised learning, what is typically provided to the algorithm during training? A)
Input data only B) Output labels only C) Both input-output pairs D) None of the above
3. Which machine learning algorithm is commonly used for classification tasks? A) Linear
regression B) K-means clustering C) Decision trees D) Principal Component Analysis
(PCA)
4. What is the primary goal of unsupervised learning? A) Minimize prediction errors B)
Classify data into predefined categories C) Discover patterns or structure in data D)
Optimize algorithm performance
5. Which evaluation metric is commonly used for regression tasks? A) Accuracy B) F1
score C) Mean Squared Error (MSE) D) Precision
6. Which of the following is not a preprocessing step in machine learning? A) Feature
scaling B) Feature extraction C) Model training D) Data cleaning
7. Which technique is used to reduce the dimensionality of data? A) Gradient descent B)
Regularization C) PCA (Principal Component Analysis) D) Stochastic gradient descent
8. Which algorithm is suitable for dealing with a large number of features and sparse
datasets? A) Support Vector Machines (SVM) B) Naive Bayes C) Random Forest D) K-
nearest neighbors (KNN)
9. What is the purpose of cross-validation in machine learning? A) To test the model on
unseen data B) To tune hyperparameters C) To preprocess data D) To evaluate model
performance
10. What does the term "overfitting" refer to in machine learning? A) Model performs well
on training data but poorly on test data B) Model generalizes well to unseen data C)
Model is too simple to capture underlying patterns D) Model is biased towards a
particular feature
11. Which algorithm is often used for anomaly detection in data? A) K-means clustering B)
Decision trees C) One-class SVM (Support Vector Machine) D) Logistic regression
12. Which technique is used to handle missing data in a dataset? A) Feature scaling B)
Principal Component Analysis (PCA) C) Imputation D) Regularization
13. Which of the following is not a type of ensemble learning technique? A) Bagging B)
Boosting C) Stacking D) Gradient descent
14. What is the activation function commonly used in neural networks for binary
classification tasks? A) Sigmoid B) ReLU (Rectified Linear Unit) C) Tanh (Hyperbolic
tangent) D) Softmax
15. Which method is used to reduce the variance of a machine learning model? A)
Regularization B) Feature scaling C) Hyperparameter tuning D) Gradient descent

Page 2 of 5
16. Which evaluation metric is suitable for imbalanced classification problems? A) Accuracy
B) Precision C) F1 score D) Mean Absolute Error (MAE)
17. In reinforcement learning, what does an agent learn from interactions with an
environment? A) Supervised patterns B) Unsupervised patterns C) Rewards and penalties
D) Feature representations
18. Which technique is used to improve model performance by penalizing large coefficients?
A) Feature scaling B) Regularization C) Principal Component Analysis (PCA) D)
Imputation
19. Which algorithm is suitable for predicting categories or classes based on historical data?
A) Linear regression B) K-means clustering C) Decision trees D) None of the above
20. Which method is used to evaluate the importance of different features in a machine
learning model? A) Gradient descent B) Regularization C) PCA (Principal Component
Analysis) D) Feature importance analysis

Answers:

1. D) Structured learning
2. C) Both input-output pairs
3. C) Decision trees
4. C) Discover patterns or structure in data
5. C) Mean Squared Error (MSE)
6. C) Model training
7. C) PCA (Principal Component Analysis)
8. A) Support Vector Machines (SVM)
9. D) To evaluate model performance
10. A) Model performs well on training data but poorly on test data
11. C) One-class SVM (Support Vector Machine)
12. C) Imputation
13. D) Gradient descent
14. A) Sigmoid
15. A) Regularization
16. C) F1 score
17. C) Rewards and penalties
18. B) Regularization
19. C) Decision trees
20. D) Feature importance analysis

SECTION B (20 MARKS)


Answer all questions in this section. Write your answers in the spaces provided after each part question.

1. What is the process of adjusting model parameters to minimize error?


2. What type of learning uses input-output pairs for training?
3. What metric measures how well a model generalizes to new data?
4. What technique reduces overfitting by adding a penalty to large coefficients?
5. What type of learning does not require labeled data for training?
6. What algorithm finds patterns in data without predefined categories?
7. What method reduces dimensionality by combining features?

Page 3 of 5
8. What function converts input signals into output predictions?
9. What type of algorithm learns from rewards and penalties?
10. What method groups data points into clusters?
11. What term describes the ability of a model to learn without human intervention?
12. What technique transforms input data to have zero mean and unit variance?
13. What type of algorithm aims to predict continuous values?
14. What metric evaluates the correctness of positive predictions?
15. What method evaluates model performance using held-out data?
16. What approach combines predictions from multiple models?
17. What concept measures the uncertainty or confidence in a model's predictions?
18. What method learns by adjusting model parameters in response to errors?
19. What type of model does not involve explicit rules or logic?
20. What technique prevents the model from memorizing training examples?

Answers:

1. Optimization
2. Supervised
3. Generalization
4. Regularization
5. Unsupervised
6. Clustering
7. Dimensionality reduction
8. Activation
9. Reinforcement
10. Clustering
11. Autonomy
12. Standardization
13. Regression
14. Precision
15. Validation
16. Ensemble
17. Uncertainty
18. Learning
19. Neural network

SECTION C (20 MARKS)


1. How does feature selection contribute to improving model performance?

Page 4 of 5
2. Explain the difference between bias and variance in machine learning models.
3. What role does the learning rate play in gradient descent optimization?
4. Describe the process of cross-validation and its importance in model evaluation.
5. How does regularization prevent overfitting in machine learning?
6. What are the advantages and disadvantages of deep learning compared to traditional machine
learning algorithms?
7. How can imbalanced datasets affect the performance of a classification model?
8. Explain the concept of hyperparameter tuning in machine learning.
9. What are the ethical considerations involved in deploying machine learning models in real-
world applications?
10. Describe the steps involved in preprocessing textual data for natural language processing
tasks.
11. How does ensemble learning improve the predictive performance of machine learning
models?
12. What are the different types of kernel functions used in Support Vector Machines (SVM)?
13. How does the choice of evaluation metric impact model selection in machine learning?
14. Explain the trade-off between bias and variance when choosing model complexity.
15. What are the key challenges in deploying machine learning models in production
environments?
16. Describe the process of feature engineering and its significance in machine learning.
17. How does the curse of dimensionality affect the performance of machine learning
algorithms?
18. What are the main differences between batch learning and online learning approaches in
machine learning?
19. Discuss the concept of transfer learning and its applications in machine learning.
20. How can interpretability and transparency be ensured in complex machine learning models?

Page 5 of 5

You might also like