UNIT 1

In Supervised Learning, the goal is to

• A. Find hidden patterns in data without labels.

• B. Learn from labeled data to make predictions.

• C. Interact with an environment to maximize a reward.

• D. None of the above.

Clustering algorithms are primarily associated with:

• A. Supervised Learning

• B. Unsupervised Learning

• C. Reinforcement Learning

• D. None of the above

When gathering data for a machine learning project, which of the following should be considered?

• A. Data quantity and quality

• B. Data relevance to the problem

• C. Labeling and annotations

• D. All of the above

Which of these algorithms is typically used in Unsupervised Learning?

• A. Decision Tree

• B. Logistic Regression

• C. K-Means Clustering

• D. Random Forest

2. A hypothesis in machine learning is:

• A. A random guess about the outcome

• B. A mathematical formula that makes predictions based on input features

• C. The true labels of the dataset

• D. An evaluation metric used in model selection

3. Which of the following can help mitigate overfitting?

• A. Increasing the model complexity

• B. Reducing the size of the training dataset

• C. Using regularization techniques

• D. Ignoring validation data

2. The confusion matrix is a tool primarily used to evaluate

• A. Regression models

• B. Classification models

• C. Clustering models

• D. Anomaly detection models

5. Recall is defined as

• A. The ability of a model to correctly identify all positive instances

• B. The accuracy of the model on all test samples

• C. The proportion of false positives in predictions

• D. The average prediction error

Which of these is a common issue when using accuracy as the sole evaluation metric?

• A. Accuracy doesn’t account for false positives and false negatives, especially on imbalanced datasets

• B. Accuracy is always low on balanced datasets

• C. Accuracy considers only true positives and ignores true negatives

• D. Accuracy increases with increasing dataset size

 A high bias error in a model typically indicates

• A. Underfitting

• B. Overfitting

• C. High model variance

• D. An optimal fit