Skip to content

Automatic Choice of EPS #87

Description

@mhahsler

Implement some of the approaches for automatic choice of the eps parameter.

  • AutoEpsDBSCAN: Computes a smoothed k-distance graph to automatically detect the "knee," suitable for large, noisy datasets.
  • GA-DBSCAN: Combines Genetic Algorithms (GA) with DBSCAN to automatically find the optimal eps by maximizing cluster accuracy, addressing the difficulty of setting this parameter in varying densities.
  • CVR-DBSCAN: Uses curvature analysis on the data to estimate appropriate eps and MinPts parameters.
  • KNN-Probability: Uses a probability-based approach (often using Gaussian distributions) to dynamically determine eps for each data frame, particularly for point cloud data.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions