Skip to content

yenchic/density_ranking

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 

Repository files navigation

Density Ranking

Analyzing data using density ranking.

  • Paper reference:
    • Chen, Yen-Chi, and Adrian Dobra. "Measuring Human Activity Spaces With Density Ranking Based on GPS Data." arXiv preprint arXiv:1708.05017 (2017).
    • Chen, Yen-Chi. "Generalized Cluster Trees and Singular Measures." arXiv preprint arXiv:1611.02762 (2016).
  • Maintianer: yenchic@uw.edu
  • DR.R: the main script for running the program.
  • DR_demo.R: an example for applying DR method to a dataset.

To try this script, please execute the programs in DR_demo.R. Note that you need to download the dataset from https://archive.ics.uci.edu/ml/datasets/GPS+Trajectories before implementing DR_demo.R.

DR.R

Here is a description for the main function.

DR

DR = function(data, h, kernel="Gaussian", xlim=NULL, ylim=NULL, CL_lev = NULL, n_res=201, n_tg=100, ...)

  • Density ranking analysis for 2D datasets (GPS).
  • Inputs:
    • data: Input 2D data matrix.
    • h: Smoothing parameter.
    • kernel: kernel function, choosing from "Gaussian", "uniform", and "quartic". Default is Gaussian kernel.
    • xlim: The range of first variable.
    • ylim: The range of second variable.
    • CL_lev: The levels for contours lines. If not specify, it will automatically choose 0.05, 0.10, ..., 0.95 as the levels.
    • n_res: The resolution of grids. 201 means a 201 by 201 matrix over the range of xlim and ylim.
    • n_tg: The levels for evaluating the mass-volume curve, Betti number, and persistence curves.
  • Outputs:
    • The density ranking analysis result; a list consisting of
      • h: The smoothing bandwidth.
      • x_grid: The grid point for the first variable.
      • y_grid: The grid point for the second variable.
      • gr_alpha: The density ranking value at each grid point.
      • data_alpha: The density ranking value at each data point.
      • CL_lev: The contour levels.
      • CL: A list of contour lines corresponding to CL_lev.
      • persistent: The persistent diagram of density ranking.
      • clevel: The levels where the mass-volume, Betti number, and persistence curves are evaluated.
      • Mcurve: The value of the mass-volume curve evaluated at clevel.
      • Bcurve: The value of the Betti number curve evaluated at clevel.
      • Pcurve: The value of the persistence curve evaluated at clevel.

Releases

No releases published

Packages

No packages published

Languages