Exploring Machine Learning Concepts

What is the primary goal of supervised learning?

• To cluster similar data points into groups.

• To predict future outcomes without any labeled data.

• To learn a mapping from input features to output labels using labeled data.

• To reduce the dimensionality of input features.

Which of the following is an example of unsupervised learning?

• Neural Networks

• Classification

• Clustering

• Regression

What is a common activation function used in neural networks?

• Softmax

• Tanh

• Sigmoid

• ReLU (Rectified Linear Unit)

How do decision trees make predictions?

• Decision trees make predictions by averaging all feature values.

• Decision trees predict outcomes based solely on the last feature value.

• Decision trees use random sampling of data points to make predictions.

• Decision trees make predictions by following decision rules from the root to a leaf node based on feature values.

What type of data is typically used in supervised learning?

• Structured data

• Unsupervised data

• Unlabeled data

• Labeled data

What is the difference between classification and regression in supervised learning?

• Classification is used for time series analysis; regression is used for clustering.

• Classification predicts future values; regression predicts past values.

• Classification deals with categorical outcomes; regression deals with continuous outcomes.

• Classification uses numerical data; regression uses text data.

What is clustering in the context of unsupervised learning?

• Clustering is the process of grouping similar data points in unsupervised learning.

• Clustering is the process of sorting data points in ascending order.

• Clustering is a method for supervised learning.

• Clustering involves labeling data points with predefined categories.

What is overfitting in neural networks?

• Overfitting is when the model performs poorly on both training and test data due to lack of data.

• Overfitting happens when a model is trained on too little data, leading to underperformance.

• Overfitting in neural networks is when the model performs well on training data but poorly on test data due to excessive complexity.

• Overfitting occurs when a model is too simple and cannot capture the underlying patterns.

What are the advantages of using decision trees?

• Limited to numerical data only

• High computational cost

• Advantages of using decision trees include interpretability, minimal data preprocessing, ability to handle various data types, and clear visualization of decisions.

• Requires extensive data cleaning

How can neural networks be trained effectively?

• Training without any data preprocessing

• Effective training of neural networks involves data preparation, architecture selection, loss function and optimizer choice, regularization, training process implementation, hyperparameter tuning, and evaluation.

• Ignoring the choice of activation functions

• Using only a single layer for all tasks