MCQs

Disclaimer:

Educational Use Only:

The website links provided are to be used strictly for educational and knowledge-gaining purposes only.

No Misuse:

Do not misuse the links or the information contained therein for any commercial, illegal, or unethical activities.

Respect for Websites and Businesses:

Respect the intellectual property and business interests of the websites and their owners. Do not attempt to alter, copy, or redistribute content without proper authorization.

Acknowledge Value of Knowledge:

Recognize and value the knowledge and resources shared by these websites. Use the information responsibly and ethically.

No Endorsement:

The provision of these links does not imply endorsement of the linked sites, their content, or their products and services. Users are responsible for evaluating the suitability and accuracy of the information.

User Responsibility:

By accessing these links, you agree to use the information at your own risk and to comply with all applicable laws and website terms of use.

Overall MCQs for Reference

Which algorithm is most suitable for predicting housing prices?
a. K-Means
b. Linear Regression
c. Decision Tree Classifier
d. Naïve Bayes
Answer: b. Linear Regressionsanfoundry+1
What is the primary advantage of ensemble learning?
a. Less computation
b. Reduced flexibility
c. Improved model accuracy
d. Limited representation
Answer: c. Improved model accuracysanfoundry
In machine learning, which of the following splits a dataset into training and test sets?
a. data_split()
b. train_test_split()
c. split_dataset()
d. train_data()
Answer: b. train_test_split()placementpreparation
Which type of task is best suited for clustering algorithms?
a. Labeled data prediction
b. Classification
c. Unlabeled data grouping
d. Regression
Answer: c. Unlabeled data groupinginterviewkickstart
Overfitting occurs when:
a. Model fits only training data well
b. Model generalizes perfectly
c. Model ignores all patterns
d. Model uses less data
Answer: a. Model fits only training data wellplacementpreparation
What is the last step in a typical machine learning pipeline?
a. Exploratory Data Analysis
b. Model Evaluation
c. Data Cleaning
d. Feature Engineering
Answer: b. Model Evaluationgeeksforgeeks
Which of these is NOT an application of machine learning?
a. Image recognition
b. Predicting stock market trends
c. Conducting manual calculations
d. Natural language processing
Answer: c. Conducting manual calculationsyoutube
Which optimization algorithm is widely used for minimizing cost functions in ML?
a. Genetic Algorithm
b. Stochastic Gradient Descent
c. Decision Tree
d. K-Means
Answer: b. Stochastic Gradient Descentsanfoundry
Which ML algorithm can handle unlabelled data efficiently?
a. Logistic Regression
b. SVM
c. Clustering
d. Decision Tree
Answer: c. Clusteringinterviewkickstart
Which of the following is the purpose of feature scaling?
a. Remove outliers
b. Speed up convergence
c. Reduce dataset size
d. Increase bias
Answer: b. Speed up convergencegeeksforgeeks
What is principal component analysis (PCA) commonly used for?
a. Feature Selection
b. Feature Extraction
c. Model Tuning
d. Data Visualization
Answer: b. Feature Extractiongeeksforgeeks
Which is a common technique for imbalanced classification tasks?
a. K-Fold Cross Validation
b. Random Oversampling
c. Underfitting
d. Dropout Regularization
Answer: b. Random Oversamplinggeeksforgeeks
What does AUC-ROC curve evaluate in ML classification models?
a. Model Complexity
b. Precision and Recall
c. Model Performance
d. Overfitting
Answer: c. Model Performancegeeksforgeeks
Which of the following is a supervised learning model?
a. K-Means
b. PCA
c. Linear Regression
d. DBSCAN
Answer: c. Linear Regressionplacementpreparation
Which function measures error in regression models?
a. Accuracy
b. Entropy
c. Mean Squared Error
d. Gini Index
Answer: c. Mean Squared Errorgeeksforgeeks
Which method is often used for feature selection?
a. Bagging
b. Lasso (L1 Regularization)
c. Boosting
d. Clustering
Answer: b. Lasso (L1 Regularization)geeksforgeeks
A confusion matrix is used to evaluate:
a. Regression models
b. Clustering results
c. Classification models
d. Feature selection
Answer: c. Classification modelsgeeksforgeeks
What is a potential issue in high-dimensional datasets?
a. Overfitting
b. Underfitting
c. Low noise
d. Reduced flexibility
Answer: a. Overfittingcompphysics.github
Which method combines predictions from several models?
a. Boosting
b. Bagging
c. Stacking
d. All of the above
Answer: d. All of the abovecompphysics.github+1
Cross-validation is typically used to:
a. Reduce data size
b. Increase training accuracy
c. Assess model generalizability
d. Add more features
Answer: c. Assess model generalizabilitysanfoundry+1
K-Nearest Neighbors (KNN) is best described as:
a. Parametric
b. Non-parametric
c. Randomized
d. Linear
Answer: b. Non-parametricsanfoundry+1
The main goal of regularization in machine learning is to:
a. Increase complexity
b. Penalize large weights
c. Reduce bias
d. Maximize accuracy
Answer: b. Penalize large weightsgeeksforgeeks
Dropout is a regularization technique mainly for:
a. Decision Trees
b. Linear Regression
c. Neural Networks
d. K-Means
Answer: c. Neural Networkscompphysics.github+1
What data split ratio is most common for train-test splitting?
a. 10-90
b. 70-30
c. 90-10
d. 50-50
Answer: b. 70-30geeksforgeeks
Bias in a model is typically caused by:
a. Too little data
b. Highly complex model
c. Too simple model
d. More features
Answer: c. Too simple modelplacementpreparation+1
Which metric assesses regression prediction quality?
a. F1-score
b. Accuracy
c. R-squared
d. Purity
Answer: c. R-squaredgeeksforgeeks
SVMs are best known for:
a. Learning decision boundaries
b. Handling categorical targets
c. Imputing missing values
d. Visualization
Answer: a. Learning decision boundariessanfoundry+1
Support Vector Machines primarily rely on:
a. Kernel functions
b. Tree splits
c. Random forests
d. Polynomial regression
Answer: a. Kernel functionssanfoundry+1
One-hot encoding:
a. Reduces model variance
b. Transforms categorical into binary
c. Trims outliers
d. Normalizes features
Answer: b. Transforms categorical into binaryplacementpreparation+1
The main goal of data preprocessing is to:
a. Visualize data
b. Improve model input quality
c. Plot graphs
d. Identify feature relevance
Answer: b. Improve model input qualitygeeksforgeeks
Ensemble methods help in:
a. Reducing model diversity
b. Increasing overfitting
c. Improving accuracy and robustness
d. Feature reduction
Answer: c. Improving accuracy and robustnesssanfoundry
What is stratified sampling used for during cross-validation?
a. Faster calculation
b. Equal representation of classes
c. Reducing sample size
d. Adding more features
Answer: b. Equal representation of classesgeeksforgeeks
Which method is NOT a dimensionality reduction?
a. PCA
b. t-SNE
c. Lasso Regression
d. Bagging
Answer: d. Baggingcompphysics.github+1
In time-series forecasting, which method accounts for seasonality?
a. K-Means
b. ARIMA
c. Linear Regression
d. Lasso
Answer: b. ARIMAcompphysics.github+1
Accuracy is a poor metric when:
a. Datasets are small
b. Data is linear
c. Classes are imbalanced
d. Overfitting is present
Answer: c. Classes are imbalancedgeeksforgeeks
Which learning method best handles continuous reward feedback?
a. Supervised Learning
b. Unsupervised Learning
c. Reinforcement Learning
d. Semi-supervised Learning
Answer: c. Reinforcement Learningcompphysics.github
Decision trees can be prone to:
a. Underfitting
b. High interpretability
c. Overfitting
d. Only regression
Answer: c. Overfittingcompphysics.github+1
What does “bagging” stand for?
a. Bootstrap aggregating
b. Baggage tagging
c. Balanced aggregation
d. Binary aggregation
Answer: a. Bootstrap aggregatingcompphysics.github+1
ROC curve plots:
a. Precision vs Recall
b. True Positive Rate vs False Positive Rate
c. Accuracy vs Error
d. Loss vs Epoch
Answer: b. True Positive Rate vs False Positive Rategeeksforgeeks
Regularization prevents:
a. Underfitting
b. Overfitting
c. Model pruning
d. Multicollinearity
Answer: b. Overfittinggeeksforgeeks
Which is a disadvantage of deep learning?
a. High interpretability
b. Scalability
c. Requires lots of data & computation
d. Handles unstructured data
Answer: c. Requires lots of data & computationcompphysics.github
What is feature engineering?
a. Extracting new inputs from raw data
b. Dropping irrelevant features
c. Model scoring
d. Model deployment
Answer: a. Extracting new inputs from raw datageeksforgeeks
The elbow method is most often used for:
a. Deciding tree depth
b. Choosing clusters (K) in K-Means
c. Regularization
d. Ensemble learning
Answer: b. Choosing clusters (K) in K-Meansplacementpreparation
In a confusion matrix, which value indicates false positives?
a. Top left
b. Top right
c. Bottom left
d. Bottom right
Answer: b. Top rightgeeksforgeeks
Which algorithm can best handle nonlinear relationships?
a. Linear Regression
b. Decision Trees
c. Logistic Regression
d. PCA
Answer: b. Decision Treessanfoundry
What is the primary focus in applied analytics in machine learning?
a. Theoretical derivation
b. Business value and actionability
c. Training error
d. Hyperparameter tuning
Answer: b. Business value and actionabilitysps.columbia+1
Which method combines weak learners into a strong one?
a. Bagging
b. Boosting
c. Dropout
d. PCA
Answer: b. Boostingsanfoundry
Imputation is the process of:
a. Scaling features
b. Filling missing values
c. Training models
d. Visualizing data
Answer: b. Filling missing valuesgeeksforgeeks
Which is most closely related to anomaly detection?
a. Outlier analysis
b. Cross-validation
c. Training error
d. Hyperparameter tuning
Answer: a. Outlier analysisplacementpreparation+1
Transfer learning is often used in:
a. Regression only
b. Large labeled datasets
c. Adapting pre-trained models to new tasks
d. Model tuning
Answer: c. Adapting pre-trained models to new taskscompphysics.github+1

Blog

Systemethic Technologies

Leave a Reply Cancel reply

Useful links

Other Terms & Policies

Contact Info