AI 용어집
인공지능 완전 사전
K-means clustering
Partitioning algorithm that groups data into K clusters by minimizing intra-cluster variance.
Hierarchical Clustering
Method that builds a hierarchy of clusters through bottom-up (agglomerative) or top-down (divisive) approaches.
DBSCAN Clustering
Density-based algorithm that identifies arbitrarily shaped clusters by grouping points that are close enough together
Principal Component Analysis (PCA)
Linear dimensionality reduction technique that transforms data into orthogonal axes maximizing variance.
t-SNE
Non-linear dimensionality reduction algorithm preserving local structures for data visualization.
Autoencoders
Unsupervised neural networks that learn efficient compressed representations through reconstruction.
Anomaly Detection
Identification of observations or patterns that deviate significantly from the normal behavior of the data.
Gaussian Mixture Models (GMM)
Probabilistic approach modeling data as a combination of multiple Gaussian distributions for clustering.
Self-Organizing Maps (SOM)
Competitive neural networks projecting high-dimensional data onto a 2D grid while preserving topology.
Non-negative Matrix Factorization (NMF)
Method that decomposes a matrix into non-negative factors to discover latent features.
Spectral Clustering
Technique using eigenvalues of a similarity matrix to perform clustering on non-convex data.
UMAP
Modern dimensionality reduction algorithm that simultaneously preserves local and global data structures.
Unsupervised GANs
Generative Adversarial Networks learning without labels to generate realistic data through competition.
Association Rules
Methods for discovering relationships between variables in large databases, such as the Apriori algorithm.
Estimation de Densité par Noyau (KDE)
Technique non paramétrique estimant la fonction de densité de probabilité d'une variable aléatoire.