[Centrality Measures]

graph-tool
1

Hi Tiago,

I'm working with some centrality measures provide by graph_tool and I have
some questions:

*Does the 'trust_map' parameter of the eigentrust centrality function refer
to the edges' weight?*
In the example present in the docs, it was used the edges' weight as value
to 'trust_map', but in the parameters' description was said that this
parameter refers to 'Edge property map with the values of trust associated
with each edge. The values must lie in the range [0,1]."

*Why are my values of authority and hub centrality equal?*
I'm getting the same values for authority and hub centrality.
ex:
v_id hub authority
1, 0.401326450573, 0.401326450573
2, 0.614525497348, 0.614525497348
...

*Code:g.vp.authority = g.new_vertex_property("double")g.vp.hub =
g.new_vertex_property("double")eig, g.vp.authority, g.vp.hub =
graph_tool.centrality.hits(g, weight = g.ep.edge_weight)*

Thais Almeida.

attachment.html (1.27 KB)

2

Hi Tiago,

I'm working with some centrality measures provide by graph_tool and I have
some questions:

*Does the 'trust_map' parameter of the eigentrust centrality function refer
to the edges' weight?*
In the example present in the docs, it was used the edges' weight as value
to 'trust_map', but in the parameters' description was said that this
parameter refers to 'Edge property map with the values of trust associated
with each edge. The values must lie in the range [0,1]."

The "edge weight" is synonymous with "edge trust" in this algorithm.

*Why are my values of authority and hub centrality equal?*
I'm getting the same values for authority and hub centrality.
ex:
v_id hub authority
1, 0.401326450573, 0.401326450573
2, 0.614525497348, 0.614525497348
...

How I am supposed to know, without looking at your graph? Is your graph even
directed?

Best,
Tiago