10 Questions of Machine Learning

What is the primary goal of supervised learning?

• Clustering

• • Classification

• • Dimensionality Reduction

• • Reinforcement Learning

Which algorithm is commonly used for linear regression in machine learning?

• k-Nearest Neighbors

• Decision Trees

• Support Vector Machines

• Linear Regression

What is overfitting in machine learning?

• Model too simple

• Model fits the training data too closely

• Lack of training data

• Model underperforms on test data

Which evaluation metric is commonly used for classification problems?

• Mean Squared Error (MSE)

• R-squared

• F1 Score

• Mean Absolute Error (MAE)

What is the role of the confusion matrix in machine learning?

• Evaluate model accuracy

• Assess the impact of feature scaling

• Visualize decision boundaries

• Summarize the performance of a classification algorithm

Which algorithm is suitable for handling imbalanced datasets in classification problems?

• • Decision Trees

• • k-Nearest Neighbors

• Support Vector Machines

• SMOTE (Synthetic Minority Over-sampling Technique)

What role does vectorization play in machine learning feature engineering?

• It helps in feature scaling

• It transforms raw data into a format suitable for input to machine learning models

• It focuses on reducing dimensionality of feature space

• It simplifies complex mathematical operations

What is the main advantage of using decision trees in machine learning?

• They are resistant to overfitting

• They require minimal preprocessing of data

• They can handle both numerical and categorical data

• They are only suitable for binary classification

What is pruning in the context of decision trees?

• Removing outliers from the dataset

• Adding more branches to increase model complexity

• Trimming branches of the tree to prevent overfitting

• The process of selecting the root node of the tree

In the context of machine learning, what potential issue does data leakage refer to during the training and testing phase?

• The model's inability to learn from the training set

• Including information in the testing set that the model learned from the training set

• The presence of outliers in the dataset

• Using too small of a training set