Machine-Learning Interview Questions

Machine Learning is a branch of AI that enables computers to learn from data without explicit programming.

Main types include Supervised Learning (labeled data), Unsupervised Learning (unlabeled data), and Reinforcement Learning (learning by trial and error).

ML is used for prediction, classification, clustering, recommendation systems, and natural language processing.

It powers applications like fraud detection, recommendation engines, chatbots, and autonomous vehicles.

Overfitting occurs when a model learns training data too well, capturing noise and irrelevant patterns, leading to poor generalization on unseen data.

Signs include high accuracy on training data but low on test data.

Prevention techniques include regularization, cross-validation, pruning, early stopping, and increasing training data.

Hyperparameter tuning and ensemble methods like Random Forests and Gradient Boosting also help mitigate overfitting.

A confusion matrix is a table used to evaluate the performance of a classification algorithm.

It shows actual vs predicted labels, with True Positives, False Positives, True Negatives, and False Negatives.

From the matrix, metrics like accuracy, precision, recall, F1-score, and ROC-AUC are derived.

Confusion matrices are essential for evaluating classifier performance and choosing the right evaluation metric based on the problem (e.g., medical diagnosis vs spam detection).