This preview shows page 1. Sign up to view the full content.
Unformatted text preview: like)
component is composed by the node initially reached, plus k other treelike components,
which have the same size distribution, where k is the number of outgoing links of the
node, whose distributionFollowing a random edge
is qk . The probability that the global component Q S has size S
is thus
(A2.6)
Q S likely to follow components a size S − of
• k times more = qk Prob(union of kedge to has node 1) degree k than a
k
node of degree 1:
(counting the initially reached node in S ). The generating function H1 is by deﬁnition
qk = kPk / k kPk .
H1 (x ) = QS x S, (A2.7) S • There are k − 1 other edges outgoing from this node.
k and the distribution of the sum of the sizes of the k components is generated by H1 (as
previously explained for the sum of degrees), i.e. • Each one of those k − 1 edges has probability qk of leading to
Prob(union of k components has size S ) · x S = ( H1 (x ))k .
(A2.8)
node of degree k .
S
A B (Circles denote isolated nodes, squares components of unknown size.) For convenience, deﬁne the GF for random edge following
(Build up more complex from simpler)...
View
Full
Document
This document was uploaded on 03/12/2014 for the course CSCI 289 at UC Davis.
 Winter '11
 RaissaD'Souza

Click to edit the document details