def top(graph,k,centrality,confidence):
if centrality == "b":
b = betweenness(graph)
elif centrality == "e":
b = eigenvector(graph,float(confidence))
top = []
top_values = []
for i in b.keys():
added = 0
for j in range(min(len(top),int(k))):
if b[top[j]] < b[i]:
top.insert(j,i)
added = 1
break
if added == 0:
top.append(i)
if len(top) > int(k):
top.pop()
for i in range(len(top)):
top_values.append(b[top[i]])
return top, top_values
def top(graph, k, centrality, confidence):
if centrality == "b":
b = betweenness(graph)
elif centrality == "e":
b = eigenvector(graph, float(confidence))
top = []
top_values = []
for i in b.keys():
added = 0
for j in range(min(len(top), int(k))):
if b[top[j]] < b[i]:
top.insert(j, i)
added = 1
break
if added == 0:
top.append(i)
if len(top) > int(k):
top.pop()
for i in range(len(top)):
top_values.append(b[top[i]])
return top, top_values
print '# Page Rank execution...'
pagerank, iterations, err = pr(g, alpha=1.0e-5, eps=1.0e-3)
print '#', iterations, ' iterations. Error:', err
top_pr = [a for a, b in topk(pagerank, seed)]
# In[6]:
print '# Eigenvector Centrality...',
cscores, diffsum = ec(g)
top_eigenc = [a for a, b in topk(cscores, seed)]
print '# Done'
# In[7]:
print '# Betweennes centrality...',
bet = betweenness(g)
top_bet = [a for a, b in topk(bet, seed)]
print '# Done'
sys.stdout.flush()
# ## Execute Independent Cascade Model
# In[8]:
seed = 100
max_pr_ad = independent_cascade(g, top_pr[:seed], 'Page Rank')
max_eigenc_ad = independent_cascade(g, top_eigenc[:seed], 'Eigenvector')
max_bet_ad = independent_cascade(g, top_bet[:seed], 'Betweenness')
pr_max_seed = seed
eigenc_max_seed = seed
print '#',iterations, ' iterations. Error:', err
top_pr = [a for a,b in topk(pagerank, nodes)]
# In[9]:
print '# Eigenvector Centrality...',
cscores, diffsum = ec(g)
top_eigenc = [a for a, b in topk(cscores, nodes)]
print '# Done'
# In[10]:
print '# Betweennes centrality...',
bet = betweenness(g)
top_bet = [a for a, b in topk(bet, nodes)]
print '# Done'
sys.stdout.flush()
# ## Execute Independent Cascade Model
seed = [i for i in xrange(200, nodes - 100, 200)]
print '# Page Rank\tEigenvector\tBetweenness'
i = 0
while True:
max_pr_ad = independent_cascade(g, top_pr[:seed[i]], 'Page Rank')
max_eigenc_ad = independent_cascade(g, top_eigenc[:seed[i]], 'Eigenvector')
max_bet_ad = independent_cascade(g, top_bet[:seed[i]], 'Betweenness')
#!/usr/bin/python
import sys
from lesson3 import readGraph, count2Paths, diameter
from lesson4 import betweenness, cb_max, topk
simple_graph = {}
simple_graph[0] = {1}
simple_graph[1] = {0, 2}
simple_graph[2] = {1, 3}
simple_graph[3] = {2, 4}
simple_graph[4] = {3}
cb = betweenness(simple_graph, True)
# graph = readGraph(sys.argv[1])
# cb = betweenness(graph)
k = 10
print 'Top ', k, '=>', topk(cb, k)
#!/usr/bin/python
import sys
from lesson1 import randomGraph
from lesson4 import betweenness
from lesson5 import eigenvector_centrality
# nodes = int(sys.argv[1]) #number of nodes
# sep = float(sys.argv[2]) #experiments for p = sep, 2sep, 3sep, ..., 1
# graph = randomGraph(nodes,sep)
simple_graph = {}
simple_graph[0] = {1}
simple_graph[1] = {0, 2}
simple_graph[2] = {1, 3}
simple_graph[3] = {2, 4}
simple_graph[4] = {3}
print betweenness(simple_graph)
print eigenvector_centrality(simple_graph)