D491 Introduction to Analytics - Set 5 - Part 3

Correct Answer: To encode categorical variables as binary vectors

One-hot encoding transforms categorical variables into binary vectors, making them suitable for machine learning models that require numerical input.

Correct Answer: To stop training once the validation performance starts to degrade

Early stopping halts training when the model's performance on validation data begins to degrade, preventing overfitting.

Correct Answer: Isolation forest

Isolation forest is widely used for anomaly detection because it isolates outliers by partitioning data points in a decision tree structure.

Correct Answer: XGBoost is faster to train and more scalable

XGBoost is faster and more scalable than traditional gradient boosting algorithms due to its advanced optimization techniques and parallel processing capabilities.

Correct Answer: To encourage sparsity in the model by shrinking coefficients to zero

L1 regularization (Lasso) encourages sparsity in the model by shrinking some of the coefficients to zero, effectively performing feature selection.

Correct Answer: The trade-off between true positive and false positive rates

The area under the ROC curve (AUC) measures the trade-off between the true positive rate and the false positive rate, indicating the model's classification ability.

Correct Answer: Data imputation

Data imputation is the process of replacing missing values with substituted values, commonly used to handle missing data in datasets.

Correct Answer: To maximize cumulative reward through interactions with the environment

In reinforcement learning, the agent interacts with the environment to maximize its cumulative reward by taking actions that lead to positive outcomes.

Correct Answer: Oversampling the minority class

Oversampling the minority class increases the number of instances in the underrepresented class, helping to balance the dataset and improve model performance.

Correct Answer: Recurrent Neural Network (RNN)

RNNs are commonly used for machine translation tasks because they can process and generate sequential data, making them ideal for handling language translation.