EVOLUTION-MANAGER
Edit File: power_centrality.html
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><title>R: Find Bonacich Power Centrality Scores of Network Positions</title> <meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> <link rel="stylesheet" type="text/css" href="R.css" /> </head><body> <table width="100%" summary="page for power_centrality {igraph}"><tr><td>power_centrality {igraph}</td><td style="text-align: right;">R Documentation</td></tr></table> <h2>Find Bonacich Power Centrality Scores of Network Positions</h2> <h3>Description</h3> <p><code>power_centrality</code> takes a graph (<code>dat</code>) and returns the Boncich power centralities of positions (selected by <code>nodes</code>). The decay rate for power contributions is specified by <code>exponent</code> (1 by default). </p> <h3>Usage</h3> <pre> power_centrality( graph, nodes = V(graph), loops = FALSE, exponent = 1, rescale = FALSE, tol = 1e-07, sparse = TRUE ) </pre> <h3>Arguments</h3> <table summary="R argblock"> <tr valign="top"><td><code>graph</code></td> <td> <p>the input graph.</p> </td></tr> <tr valign="top"><td><code>nodes</code></td> <td> <p>vertex sequence indicating which vertices are to be included in the calculation. By default, all vertices are included.</p> </td></tr> <tr valign="top"><td><code>loops</code></td> <td> <p>boolean indicating whether or not the diagonal should be treated as valid data. Set this true if and only if the data can contain loops. <code>loops</code> is <code>FALSE</code> by default.</p> </td></tr> <tr valign="top"><td><code>exponent</code></td> <td> <p>exponent (decay rate) for the Bonacich power centrality score; can be negative</p> </td></tr> <tr valign="top"><td><code>rescale</code></td> <td> <p>if true, centrality scores are rescaled such that they sum to 1.</p> </td></tr> <tr valign="top"><td><code>tol</code></td> <td> <p>tolerance for near-singularities during matrix inversion (see <code><a href="../../Matrix/html/solve-methods.html">solve</a></code>)</p> </td></tr> <tr valign="top"><td><code>sparse</code></td> <td> <p>Logical scalar, whether to use sparse matrices for the calculation. The ‘Matrix’ package is required for sparse matrix support</p> </td></tr> </table> <h3>Details</h3> <p>Bonacich's power centrality measure is defined by <i>C_BP(alpha,beta)=alpha (I-beta A)^-1 A 1</i>, where <i>beta</i> is an attenuation parameter (set here by <code>exponent</code>) and <i>A</i> is the graph adjacency matrix. (The coefficient <i>alpha</i> acts as a scaling parameter, and is set here (following Bonacich (1987)) such that the sum of squared scores is equal to the number of vertices. This allows 1 to be used as a reference value for the “middle” of the centrality range.) When <i>beta->1/lambda_A1</i><i>beta->1/lambda_A1</i> (the reciprocal of the largest eigenvalue of <i>A</i>), this is to within a constant multiple of the familiar eigenvector centrality score; for other values of <i>β</i>, the behavior of the measure is quite different. In particular, <i>β</i> gives positive and negative weight to even and odd walks, respectively, as can be seen from the series expansion <i>C_BP(alpha,beta) = alpha sum( beta^k A^(k+1) 1, k in 0..infinity )</i><i>C_BP(alpha,beta) = alpha sum( beta^k A^(k+1) 1, k in 0..infinity )</i> which converges so long as <i>|beta|<1/lambda_A1</i><i>|beta|<1/lambda_A1</i>. The magnitude of <i>beta</i> controls the influence of distant actors on ego's centrality score, with larger magnitudes indicating slower rates of decay. (High rates, hence, imply a greater sensitivity to edge effects.) </p> <p>Interpretively, the Bonacich power measure corresponds to the notion that the power of a vertex is recursively defined by the sum of the power of its alters. The nature of the recursion involved is then controlled by the power exponent: positive values imply that vertices become more powerful as their alters become more powerful (as occurs in cooperative relations), while negative values imply that vertices become more powerful only as their alters become <em>weaker</em> (as occurs in competitive or antagonistic relations). The magnitude of the exponent indicates the tendency of the effect to decay across long walks; higher magnitudes imply slower decay. One interesting feature of this measure is its relative instability to changes in exponent magnitude (particularly in the negative case). If your theory motivates use of this measure, you should be very careful to choose a decay parameter on a non-ad hoc basis. </p> <h3>Value</h3> <p>A vector, containing the centrality scores. </p> <h3>Warning </h3> <p>Singular adjacency matrices cause no end of headaches for this algorithm; thus, the routine may fail in certain cases. This will be fixed when I get a better algorithm. <code>power_centrality</code> will not symmetrize your data before extracting eigenvectors; don't send this routine asymmetric matrices unless you really mean to do so. </p> <h3>Note</h3> <p>This function was ported (ie. copied) from the SNA package. </p> <h3>Author(s)</h3> <p>Carter T. Butts (<a href="http://www.faculty.uci.edu/profile.cfm?faculty_id=5057">http://www.faculty.uci.edu/profile.cfm?faculty_id=5057</a>), ported to igraph by Gabor Csardi <a href="mailto:csardi.gabor@gmail.com">csardi.gabor@gmail.com</a> </p> <h3>References</h3> <p>Bonacich, P. (1972). “Factoring and Weighting Approaches to Status Scores and Clique Identification.” <em>Journal of Mathematical Sociology</em>, 2, 113-120. </p> <p>Bonacich, P. (1987). “Power and Centrality: A Family of Measures.” <em>American Journal of Sociology</em>, 92, 1170-1182. </p> <h3>See Also</h3> <p><code><a href="eigen_centrality.html">eigen_centrality</a></code> and <code><a href="alpha_centrality.html">alpha_centrality</a></code> </p> <h3>Examples</h3> <pre> # Generate some test data from Bonacich, 1987: g.c <- graph( c(1,2,1,3,2,4,3,5), dir=FALSE) g.d <- graph( c(1,2,1,3,1,4,2,5,3,6,4,7), dir=FALSE) g.e <- graph( c(1,2,1,3,1,4,2,5,2,6,3,7,3,8,4,9,4,10), dir=FALSE) g.f <- graph( c(1,2,1,3,1,4,2,5,2,6,2,7,3,8,3,9,3,10,4,11,4,12,4,13), dir=FALSE) # Compute power centrality scores for (e in seq(-0.5,.5, by=0.1)) { print(round(power_centrality(g.c, exp=e)[c(1,2,4)], 2)) } for (e in seq(-0.4,.4, by=0.1)) { print(round(power_centrality(g.d, exp=e)[c(1,2,5)], 2)) } for (e in seq(-0.4,.4, by=0.1)) { print(round(power_centrality(g.e, exp=e)[c(1,2,5)], 2)) } for (e in seq(-0.4,.4, by=0.1)) { print(round(power_centrality(g.f, exp=e)[c(1,2,5)], 2)) } </pre> <hr /><div style="text-align: center;">[Package <em>igraph</em> version 1.3.5 <a href="00Index.html">Index</a>]</div> </body></html>