ncuth

Minimum Normalised Cut Hyperplane
[IDX,SOL] = NCUTH(DATA, VARARGIN)

  [IDX,SOL] = NCUTH(X) bi-partitions the points in the N-by-D data matrix X with
  the hyperplane that minimises the normalised cut criterion (computed 
  on one-dimensional projections of the data).
 
  NCUTH returns a vector IDX containing the binary cluster assignment and a
  minimum Normalised Cut Hyperplane (nchp) object SOL. (If S initial projection
  vectors are specified S nchp hyperplanes are returned: see v0 option)

  [IDX,SOL] = NCUTH(X, 'PARAM1',val1, 'PARAM2',val2, ...) specifies optional parameters
  in the form of name/value pairs.

  OPTIONAL PARAMETERS:
  'v0' - D-by-S matrix of S initial projection vectors.
       (default: v0 = pca(X,'NumComponents',1) : First principal component of X)

  'sigma' - positive numeric scaling parameter (sigma)
    (default: sigma = 100*\sqrt(\lambda_1)*N^(-0.2), where \lambda_1 is the 1st eigenvalue of cov(X))

  'minsize' - Minimum cluster size (integer)
    (default minsize = 1)

  'maxit' - Number of BFGS iterations to perform for each value of alpha (default: 50)

  'ftol' - Stopping criterion for change in objective function value over consecutive iterations
    (default: 1.e-7)

  'verb' - Verbosity. Values greater than 0 enable visualisation during execution
    (default: 0)

  'labels' - true cluster assignment. Enables the computation of performance over 
    successive iterations and a better visualisation of how clusters are split

  'colours' - Matrix containing colour specification for observations in different clusters
    Number of rows must be equal to the number of true clusters (if 'labels' has been specified) or equal to 2.

Reference:
D.P. Hofmeyr. Clustering by Minimum Cut Hyperplanes.
IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(8):1547-1560, 2017.