Machine Learning
147 views
Feature Engineering
Quick Definition
Creating and transforming input variables to improve performance
Full Definition
Using domain knowledge to create, select, and transform input variables to improve model performance.
Examples
interaction terms, encoding categoricals, scaling features
Related Terms
data-preprocessing
feature-selection
More Machine Learning Terms
Hyperparameter Tuning
Finding optimal configuration values before training
Gaussian Mixture Model
Probabilistic model using multiple Gaussian distributions
Naive Bayes
Probabilistic classifier using Bayes theorem with independence
Data Augmentation
Artificially increasing training data through modified copies
Active Learning
Algorithm selecting most informative data points for labeling
Regularization
Techniques constraining model complexity to prevent overfitting