/usr/share/doc/mcl/html/clminfo.html is in mcl-doc 1:14-137-1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 | <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- Copyright (c) 2014 Stijn van Dongen -->
<head>
<meta name="keywords" content="manual">
<style type="text/css">
/* START aephea.base.css */
body
{ text-align: justify;
margin-left: 0%;
margin-right: 0%;
}
a:link { text-decoration: none; }
a:active { text-decoration: none; }
a:visited { text-decoration: none; }
a:link { color: #1111aa; }
a:active { color: #1111aa; }
a:visited { color: #111166; }
a.local:link { color: #11aa11; }
a.local:active { color: #11aa11; }
a.local:visited { color: #116611; }
a.intern:link { color: #1111aa; }
a.intern:active { color: #1111aa; }
a.intern:visited { color: #111166; }
a.extern:link { color: #aa1111; }
a.extern:active { color: #aa1111; }
a.extern:visited { color: #661111; }
a.quiet:link { color: black; }
a.quiet:active { color: black; }
a.quiet:visited { color: black; }
div.verbatim
{ font-family: monospace;
margin-top: 1em;
margin-bottom: 1em;
font-size: 10pt;
margin-left: 2em;
white-space: pre;
}
div.indent
{ margin-left: 8%;
margin-right: 0%;
}
.right { text-align: right; }
.left { text-align: left; }
.nowrap { white-space: nowrap; }
.item_leader
{ position: relative;
margin-left: 8%;
}
.item_compact { position: absolute; vertical-align: baseline; }
.item_cascade { position: relative; }
.item_leftalign { text-align: left; }
.item_rightalign
{ width: 2em;
text-align: right;
}
.item_compact .item_rightalign
{ position: absolute;
width: 52em;
right: -2em;
text-align: right;
}
.item_text
{ position: relative;
margin-left: 3em;
}
.smallcaps { font-size: smaller; text-transform: uppercase }
/* END aephea.base.css */
body { font-family: "Garamond", "Gill Sans", "Verdana", sans-serif; }
body
{ text-align: justify;
margin-left: 8%;
margin-right: 8%;
}
</style>
<title>The clm info manual</title>
</head>
<body>
<p style="text-align:right">
16 May 2014
<a class="local" href="clminfo.ps"><b>clm info</b></a>
14-137
</p>
<div class=" itemize " style="margin-top:1em; font-size:100%">
<div class=" item_compact"><div class=" item_rightalign nowrap " style="right:-3em">1.</div></div>
<div class=" item_text " style="margin-left:4em">
<a class="intern" href="#name">NAME</a>
</div>
<div class=" item_compact"><div class=" item_rightalign nowrap " style="right:-3em">2.</div></div>
<div class=" item_text " style="margin-left:4em">
<a class="intern" href="#synopsis">SYNOPSIS</a>
</div>
<div class=" item_compact"><div class=" item_rightalign nowrap " style="right:-3em">3.</div></div>
<div class=" item_text " style="margin-left:4em">
<a class="intern" href="#description">DESCRIPTION</a>
</div>
<div class=" item_compact"><div class=" item_rightalign nowrap " style="right:-3em">4.</div></div>
<div class=" item_text " style="margin-left:4em">
<a class="intern" href="#_section_4">OPTIONS</a>
</div>
<div class=" item_compact"><div class=" item_rightalign nowrap " style="right:-3em">5.</div></div>
<div class=" item_text " style="margin-left:4em">
<a class="intern" href="#author">AUTHOR</a>
</div>
<div class=" item_compact"><div class=" item_rightalign nowrap " style="right:-3em">6.</div></div>
<div class=" item_text " style="margin-left:4em">
<a class="intern" href="#seealso">SEE ALSO</a>
</div>
<div class=" item_compact"><div class=" item_rightalign nowrap " style="right:-3em">7.</div></div>
<div class=" item_text " style="margin-left:4em">
<a class="intern" href="#references">REFERENCES</a>
</div>
</div>
<a name="name"></a>
<h2>NAME</h2>
<p style="margin-bottom:0" class="asd_par">
clm_info — compute performance measures for graphs and clusterings.</p>
<p style="margin-bottom:0" class="asd_par">clminfo is not in actual fact a program. This manual
page documents the behaviour and options of the clm program when
invoked in mode <i>info</i>. The options <b>-h</b>, <b>--apropos</b>,
<b>--version</b>, <b>-set</b>, <b>--nop</b> are accessible
in all <b>clm</b> modes. They are described
in the <a class="local sibling" href="clm.html">clm</a> manual page.</p>
<a name="synopsis"></a>
<h2>SYNOPSIS</h2>
<p style="margin-bottom:0" class="asd_par">
<b>clm info</b> [options] <graph file> <cluster file> <cluster file>*
</p>
<p style="margin-bottom:0" class="asd_par">
<b>clm info</b>
<a class="intern" href="#opt-o"><b>[-o</b> fname (<i>write to file <b>fname</b></i>)<b>]</b></a>
<a class="intern" href="#opt-pi"><b>[-pi</b> f (<i>apply inflation beforehand</i>)<b>]</b></a>
<a class="intern" href="#opt-tf"><b>[-tf</b> spec (<i>apply tf-spec to input matrix</i>)<b>]</b></a>
<a class="intern" href="#opt-cl-tree"><b>[-cl-tree</b> fname (<i>expect file with nested clusterings</i>)<b>]</b></a>
<a class="intern" href="#opt-cat-max"><b>[-cat-max</b> num (<i>do at most <b>num</b> tree levels</i>)<b>]</b></a>
<a class="intern" href="#opt-cl-ceil"><b>[-cl-ceil</b> <num> (<i>skip clusters of size exceeding <num></i>)<b>]</b></a>
<a class="intern" href="#opt--node-self-measures"><b>[--node-self-measures</b> (<i>dump measure for native cluster</i>)<b>]</b></a>
<a class="intern" href="#opt--node-all-measures"><b>[--node-all-measures</b> (<i>dump measure for incident cluster</i>)<b>]</b></a>
<a class="intern" href="#opt-h"><b>[-h</b> (<i>print synopsis, exit</i>)<b>]</b></a>
<a class="intern" href="#opt--apropos"><b>[--apropos</b> (<i>print synopsis, exit</i>)<b>]</b></a>
<a class="intern" href="#opt--version"><b>[--version</b> (<i>print version, exit</i>)<b>]</b></a>
<matrix file> <cluster file> <cluster file>*
</p>
<a name="description"></a>
<h2>DESCRIPTION</h2>
<p style="margin-bottom:0" class="asd_par">
<b>clm info</b> computes several numbers indicative for the efficiency with
with a clustering captures the edge mass of a given graph.
Use it in conjunction with <b>clm dist</b> to determine which clusterings
you accept. See the EXAMPLES section in <b>clm dist</b>
for an example of <b>clm dist</b> and <b>clm info</b> (and <b>clm meet</b>) usage.
Output can be generated for multiple clusterings at the same time.</p>
<p style="margin-bottom:0" class="asd_par">
The <b>efficiency</b> factor is described in [1] (see
the <a class="intern" href="#references">REFERENCES</a> section). It tries to balance the dual aims of
capturing a lot of edges or edge weights and keeping the cluster footprint
or area fraction small. The efficiency number has several appealing
mathematical properties, cf. [1]. It is related to, but not derivable from,
the second and third numbers, the <i>mass fraction</i> and the
<i>area fraction</i>.</p>
<p style="margin-bottom:0" class="asd_par">
The <b>mass fraction</b> is defined as follows.
Let <b>e</b> be an edge of the graph. The clustering <i>captures</i> <b>e</b>
if the two nodes associated with <b>e</b> are in the same cluster.
Now the mass fraction is the joint weight of all captured edges divided
by the joint weight of all edges in the input graph.</p>
<p style="margin-bottom:0" class="asd_par">
The <b>area fraction</b> is roughly the sum of the
squares of all cluster sizes for all clusters in the clustering, divided by
the square of the number of nodes in the graph. It says <i>roughly</i>,
because the actual formula uses the quantity <b>N</b>*(<b>N-1</b>) wherever it
says square (of <b>N</b>) above. A low/high area fraction indicates a
fine-grained/coarse clustering.</p>
<a name="_section_4"></a>
<h2>OPTIONS</h2>
<div class=" itemize " style="margin-top:1em; font-size:100%">
<div class=" item_cascade"><div class=" item_leftalign nowrap " ><a name="opt-o"></a><b>-o</b> fname (<i>output file name</i>)</div></div>
<div class=" item_text " style="margin-left:2em">
</div>
<div style="margin-top:0em"> </div><div class=" item_cascade"><div class=" item_leftalign nowrap " ><a name="opt-pi"></a><b>-pi</b> f (<i>apply inflation beforehand</i>)</div></div>
<div class=" item_text " style="margin-left:2em">
<p style="margin-top:0em; margin-bottom:0em">
Apply inflation to the graph matrix and compute the performance
measures for the result.</p>
</div>
<div style="margin-top:0em"> </div><div class=" item_cascade"><div class=" item_leftalign nowrap " ><a name="opt-tf"></a><b>-tf</b> <tf-spec> (<i>transform input matrix values</i>)</div></div>
<div class=" item_text " style="margin-left:2em">
<p style="margin-top:0em; margin-bottom:0em">shared_defopt{-tf}</p>
</div>
<div style="margin-top:0em"> </div><div class=" item_cascade item_leftalign nowrap" ><a name="opt-cl-tree"></a><b>-cl-tree</b> fname (<i>expect file with nested clusterings (cone format)</i>)</div><div class=" item_cascade item_leftalign nowrap" ><a name="opt-cl-ceil"></a><b>-cl-ceil</b> <num> (<i>skip (nested) clusters of size exceeding <num></i>)</div>
<div class=" item_text " style="margin-left:2em">
<p style="margin-top:0em; margin-bottom:0em">
The specified file should contain a hierarchy of nested
clusterings such as generated by <b>mclcm</b>. The output is then
in a special format, undocumented but easy to understand.
Its purpose is to help cherrypick a single clustering
from a tree, in conjunction with the slightly experimental
and undocumented program <b>mlmfifofum</b>.
</p>
<p style="margin-bottom:0" class="asd_par">
The measure that is used is very slow to compute for large clusters, and
generally it will be outside any interesting range (i.e. it will be small).
Use <b>-cl-ceil</b> to skip clusters exceeding the specified size —
<b>clm info</b> will directly proceed to subclusters if they exist.
</p>
</div>
<div style="margin-top:0em"> </div><div class=" item_cascade"><div class=" item_leftalign nowrap " ><a name="opt-cat-max"></a><b>-cat-max</b> num (<i>do at most num levels</i>)</div></div>
<div class=" item_text " style="margin-left:2em">
<p style="margin-top:0em; margin-bottom:0em">
This only has effect when used with <b>-cl-tree</b>.
<b>clm info</b> will start at the most fine-grained level, working upwards.
</p>
</div>
<div style="margin-top:0em"> </div><div class=" item_cascade item_leftalign nowrap" ><a name="opt--node-all-measures"></a><b>--node-all-measures</b> (<i>dump node-wise criteria for all incident clusters</i>)</div><div class=" item_cascade item_leftalign nowrap" ><a name="opt--node-self-measures"></a><b>--node-self-measures</b> (<i>dump node-wise criteria for native cluster</i>)</div>
<div class=" item_text " style="margin-left:2em">
<p style="margin-top:0em; margin-bottom:0em">
These options return a key-value based format, with the meaning of
the keys as follows.
</p>
<div class="verbatim">nm file name (redundant unless multiple cluster files are provided)
ni node index
ci cluster index
nn number of neighbours of this node (constant for a give node)
nc cluster size (constant for a given cluster)
ef efficiency for this node/cluster combination
em max-efficiency for this node/cluster combination
mf mass fraction: percentage of edge weights for this node in this cluster
ma total mass of edge weights for this node in this cluster
xn number of neighbours of the node that are not in the cluster
xc number of nodes in the cluster that are not a neighbour of the node
ns number of neighbours of the node that are also in this cluster
ti the maximum of the edge weights for neighbours of this node that are in this cluster
to the maximum of the edge weights for neighbours of this node that are NOT in this cluster
al (alien) 1 if the node is not native to the cluster, 0 if the node is native</div>
</div>
</div>
<a name="author"></a>
<h2>AUTHOR</h2>
<p style="margin-bottom:0" class="asd_par">
Stijn van Dongen.</p>
<a name="seealso"></a>
<h2>SEE ALSO</h2>
<p style="margin-bottom:0" class="asd_par">
<a class="local sibling" href="mclfamily.html">mclfamily</a> for an overview of all the documentation
and the utilities in the mcl family.</p>
<a name="references"></a>
<h2>REFERENCES</h2>
<p style="margin-bottom:0" class="asd_par">
[1] Stijn van Dongen. <i>Performance criteria for graph clustering and Markov
cluster experiments</i>. Technical Report INS-R0012, National Research
Institute for Mathematics and Computer Science in the Netherlands,
Amsterdam, May 2000.<br>
<a class="extern" href="http://www.cwi.nl/ftp/CWIreports/INS/INS-R0012.ps.Z">http://www.cwi.nl/ftp/CWIreports/INS/INS-R0012.ps.Z</a></p>
</body>
</html>
|