Skip to content

connected-components

This command will find the connected components in a KGTK edge file. The output file is an edge file which contains the following columns:

  • node1: this column contains the nodes in the graph
  • label: this column contains only 'connected_component'
  • node2: this column contains an integer which represents the component that a node belongs to. Nodes belonging to a connected component will have the same value in this column

Usage

usage: kgtk connected-components [-h] [-i INPUT_FILE] [-o OUTPUT_FILE]
                                 [--properties PROPERTIES] [--undirected]
                                 [--strong]
                                 [--cluster-name-method {cat,hash,first,last,shortest,longest,numbered,prefixed,lowest,highest}]
                                 [--cluster-name-separator CLUSTER_NAME_SEPARATOR]
                                 [--cluster-name-prefix CLUSTER_NAME_PREFIX]
                                 [--cluster-name-zfill CLUSTER_NAME_ZFILL]
                                 [--minimum-cluster-size MINIMUM_CLUSTER_SIZE]
                                 [-v [optional True|False]]

Find all the connected components in an undirected or directed Graph.

Additional options are shown in expert help.
kgtk --expert connected-components --help

optional arguments:
  -h, --help            show this help message and exit
  -i INPUT_FILE, --input-file INPUT_FILE
                        The KGTK file to find connected components in. (May be
                        omitted or '-' for stdin.)
  -o OUTPUT_FILE, --output-file OUTPUT_FILE
                        The KGTK output file. (May be omitted or '-' for
                        stdout.)
  --properties PROPERTIES
                        A comma separated list of properties to traverse while
                        finding connected components, by default all
                        properties will be considered
  --undirected          Specify if the input graph is undirected, default
                        FALSE
  --strong              Treat graph as directed or not, independent of its
                        actual directionality.
  --cluster-name-method {cat,hash,first,last,shortest,longest,numbered,prefixed,lowest,highest}
                        Determine the naming method for clusters.
                        (default=Method.HASH)
  --cluster-name-separator CLUSTER_NAME_SEPARATOR
                        Specify the separator to be used in cat and hash
                        cluster name methods. (default=+)
  --cluster-name-prefix CLUSTER_NAME_PREFIX
                        Specify the prefix to be used in the prefixed and hash
                        cluster name methods. (default=CLUS)
  --cluster-name-zfill CLUSTER_NAME_ZFILL
                        Specify the zfill to be used in the numbered and
                        prefixed cluster name methods. (default=4)
  --minimum-cluster-size MINIMUM_CLUSTER_SIZE
                        Specify the minimum cluster size. (default=2)

  -v [optional True|False], --verbose [optional True|False]
                        Print additional progress messages (default=False).
OPTIONS:

-o {string}: Path to the output edge file.

--properties {p1, p2, ...}: Properties to consider while finding connected components. Default: All properties are considered.

--undirected: Option to specify that input file contains undirected graph.

--strong: Option to find strongly connected components (If graph is directed), or to treat graph as undirected and find connected components.

Examples

Find connected URI's that redirect to the same page

kgtk connected-components -i Dbpedia_redirects.tsv -o connected-dbpedia_uris.tsv