Algorithms For Dummies

 Exploring the World of Graphs

print('Graph A:', PageRank_teleporting(Graph_A,

                                    rounding=8))

print('Graph B:', PageRank_teleporting(Graph_B,

                                    rounding=8))

print('Graph C:', PageRank_teleporting(Graph_C,

                                    rounding=8))

Graph A: [ 0.15477863  0.15346061  0.22122243  0.15477863

           0.15787985  0.15787985]

Warning: G is substochastic

Graph B: [ 0.16502904  0.14922238  0.11627717  0.16502904

           0.20222118  0.20222118]

Graph C: [ 0.0598128   0.08523323  0.12286869  0.18996342

           0.30623677  0.23588508]

After applying the modifications to a new function, 

PageRank_teleporting

, you 

can get similar estimates for the first graph and much more realistic (and useful) 

estimates for both the second and third graphs, without falling into the traps of 

dead ends or rank sinks. Interestingly, the function is equivalent to the one 

 provided by NetworkX: 

http://networkx.readthedocs.io/en/networkx-1.11/ 

reference/generated/networkx.algorithms.link_analysis.pagerank_alg. 

pagerank.html

.

nx.pagerank(Graph_A, alpha=0.85)

{1: 0.15477892494151968,

 2: 0.1534602056628941,

 3: 0.2212224378270561,

 4: 0.15477892494151968,

 5: 0.1578797533135051,

 6: 0.15787975331350507}