Unsupervised Machine Learning

Unsupervised Machine Learning

• No labeling task

Tasks

• dimensionality reduction, reducing the number of input features in a dataset
• anomaly detection, detecting instances that are very different from the norm, also used for data cleaning
• clustering, grouping similar instances into clusters
• density estimation, estimating the density of the distribution (PDF) of data points
• association rule learning, to detect unobvious relationships between variables in a dataset

Algorithm

• Dimensionality reduction
  • Principle Component Analysis: Incremental PCA, Randomized PCA, Kernel PCA
  • Manifold Learning - LLE, Isomap, t-SNE
  • Autoencoders: Denoising AE, Variational AE, Convolutional AE, Recurrent AE
• Anomaly Detection
  • Isolation Forest
  • One class SVM
  • Local Outlier Factor
  • Minimum Covariance Determinant
• Clustering
  • K-means clustering
  • KNN (k-nearest neighbors)
  • Hierarchal clustering and Spectral Clustering
  • Affinity Propagation
  • Mean Shift and BIRCH
  • Gaussian Mixture Models
  • DBSCAN
  • Mean shift
  • BIRCH
• Density estimation
  • DBSCAN
  • Mean shift
• Association rule learning
  • Apriori
  • Eclat
  • FP-Growth

References