Machine Learning
139 views
K-Nearest Neighbors
Quick Definition
Classifying based on majority class of k nearest neighbors
Full Definition
A non-parametric algorithm classifying new points based on the majority class of their k nearest neighbors.
Examples
recommendation systems, pattern recognition, imputation
Related Terms
instance-learning
classification
clustering
More Machine Learning Terms
Random Forest
Ensemble of decision trees that votes on predictions
Lasso Regression
Regularized regression using L1 penalty for feature selection
UMAP
Fast dimensionality reduction preserving local and global structure
Autoencoder
Neural network learning compressed data representations
Confusion Matrix
Table visualizing classifier performance by class predictions
Contrastive Learning
Self-supervised learning by contrasting data pairs