Machine learning algorithms for EE – MCQs – EE

1. What is the main objective of Machine Learning (ML) in Electrical Engineering?
(A) To enable systems to learn from data and improve performance automatically
(B) To increase manual operations
(C) To replace computers with analog devices
(D) To design electrical circuits only

2. Which of the following is a supervised learning algorithm used in EE applications?
(A) Support Vector Machine (SVM)
(B) K-Means Clustering
(C) Principal Component Analysis (PCA)
(D) Genetic Algorithm

3. Unsupervised learning algorithms are mainly used for:
(A) Data clustering and pattern discovery
(B) Label-based classification
(C) Manual data entry
(D) Fault isolation

4. Which ML algorithm is commonly used for fault classification in power systems?
(A) Artificial Neural Network (ANN)
(B) Boolean logic
(C) Fourier analysis
(D) PID control

5. In supervised learning, the model is trained using:
(A) Labeled data
(B) Unlabeled data
(C) Random noise
(D) Only numeric values

6. Which of the following is an unsupervised learning technique?
(A) K-Means Clustering
(B) Decision Trees
(C) Naive Bayes
(D) Linear Regression

7. In electrical load forecasting, which algorithm provides nonlinear prediction capability?
(A) Artificial Neural Networks (ANN)
(B) Simple Linear Regression
(C) Boolean logic
(D) Relay operation

8. Which algorithm is widely used for classification problems in ML?
(A) Decision Trees
(B) Fourier Series
(C) Differential Equations
(D) Kirchhoff’s Law

9. The K-Nearest Neighbors (KNN) algorithm is based on:
(A) Distance between data points
(B) Genetic evolution
(C) Random probability
(D) Electrical resistance

10. The process of training in machine learning refers to:
(A) Adjusting model parameters based on data
(B) Hardware calibration
(C) Voltage tuning
(D) Circuit optimization

11. Which ML algorithm is suitable for dimensionality reduction?
(A) Principal Component Analysis (PCA)
(B) Logistic Regression
(C) Decision Tree
(D) Random Forest

12. In electrical engineering, regression models are often used for:
(A) Predicting continuous variables like load or temperature
(B) Identifying discrete faults only
(C) Classifying components
(D) Designing circuits

13. The Naive Bayes algorithm is based on:
(A) Bayes’ theorem and probability theory
(B) Newton’s laws
(C) Kirchhoff’s laws
(D) Boolean algebra

14. In machine learning, overfitting occurs when:
(A) The model performs well on training data but poorly on new data
(B) The model generalizes well
(C) The data is too small
(D) The model ignores outliers

15. Which algorithm can be used for energy consumption prediction in buildings?
(A) Linear Regression
(B) K-Means
(C) Clustering only
(D) PID controller

16. Random Forests are an ensemble method that use:
(A) Multiple decision trees for better accuracy
(B) Single regression model
(C) Static logic
(D) Random circuit models

17. The learning rate in an algorithm determines:
(A) How quickly a model updates its parameters
(B) The number of output nodes
(C) Data transmission speed
(D) Sensor sampling rate

18. In smart grids, machine learning helps with:
(A) Demand prediction and fault detection
(B) Manual switching
(C) Power losses
(D) Reactive power control only

19. The gradient descent method is used to:
(A) Minimize error during model training
(B) Calculate voltage
(C) Analyze circuits
(D) Detect faults manually

20. Which algorithm is suitable for anomaly detection in electrical systems?
(A) Isolation Forest
(B) Boolean logic
(C) Relay sequencing
(D) Static regression

21. The confusion matrix is used to:
(A) Evaluate classification model performance
(B) Measure voltage deviation
(C) Analyze phase shift
(D) Detect resistance values

22. Machine learning in EE is used to:
(A) Automate fault prediction, control, and energy optimization
(B) Replace sensors
(C) Increase data errors
(D) Reduce automation

23. The training dataset is used for:
(A) Building and fitting the machine learning model
(B) Testing model accuracy
(C) Generating voltage
(D) Storing results only

24. Which algorithm is commonly used for time-series forecasting in EE applications?
(A) Long Short-Term Memory (LSTM) networks
(B) K-Means Clustering
(C) Random Forest
(D) Logistic Regression

25. Cross-validation in ML helps to:
(A) Test the model’s ability to generalize to unseen data
(B) Increase data errors
(C) Simplify hardware
(D) Ignore outliers

26. Which of the following algorithms is inspired by biological evolution?
(A) Genetic Algorithm (GA)
(B) Linear Regression
(C) SVM
(D) Naive Bayes

27. In power systems, reinforcement learning (RL) can be used for:
(A) Adaptive control and energy optimization
(B) Fixed scheduling
(C) Manual fault correction
(D) Circuit simplification

28. A hyperparameter in machine learning is:
(A) A parameter set before training that controls model behavior
(B) A voltage coefficient
(C) A circuit constant
(D) A runtime variable

29. Which ML algorithm is best for load classification in smart grids?
(A) Support Vector Machine (SVM)
(B) PCA
(C) Fourier Transform
(D) Boolean function

30. The ultimate goal of applying ML algorithms in Electrical Engineering is to:
(A) Improve system efficiency, reliability, and automation through data-driven insights
(B) Increase manual computation
(C) Reduce sensor usage
(D) Simplify hardware only