Interface AlgorithmRunner

All Known Implementing Classes:
BpcRunner, FofcRunner, FtfcRunner, TscRunner
public interface AlgorithmRunner
Common interface for the four cluster-recovery algorithms used in the simulation study: TSC, BPC, FOFC, and FTFC.

Each implementation is responsible for extracting whatever input it needs (a CovarianceMatrix or the raw DataSet) from the supplied DataSet, running its search, and converting the result to a uniform List<Set<Node>>.

The effective sample size passed to algorithms that require it is always equal to the nominal sample size (dataSet.getNumRows()).

Author:
josephramsey

  • Method Summary

    Modifier and Type
    Method
    Description
    String
    label()
    Returns a short human-readable label for use in result tables.
    List<Set<Node>>
    run(DataSet dataSet)
    Runs the algorithm on dataSet and returns the recovered clusters.

  • Method Details

    • run

      List<Set<Node>> run(DataSet dataSet)
      Runs the algorithm on dataSet and returns the recovered clusters.
      Parameters:
      dataSet - the simulated data; must not be null.
      Returns:
      a list of recovered clusters, each cluster being a set of Node objects drawn from dataSet.getVariables(). Returns an empty list (never null) if the algorithm finds no clusters.

    • label

      String label()
      Returns a short human-readable label for use in result tables. Examples: "TSC", "BPC", "FOFC", "FTFC".
      Returns:
      the algorithm label; never null.