Graphics Reference
In-Depth Information
list already contains the same co-star, from a second movie), that link is
thrown away. At the end of the first pass, what remains is a tree containing
all the first-level links to Leonardo DiCaprio and a slightly shorter list of
links. This process is then repeated, using the first-level tree as the set of
nodes of interest. The process repeats for as many levels as desired.
This results in the graph shown in
Figure 10-5
.
At only two levels of
connection,LeonardoDiCaprioisconnectedtomorethan3,300actorsfrom
the original 20,000 actor data set.
Popularity (as measured by the frequency of nodes occurring from the
original Wikipedia data extract) is used to size nodes. Larger nodes tend
to occur for actors who have appeared in many films documented in the
Wikipediadataset(suchasMichaelCaine,ChristopherLee,RobertDeNiro,
Prakash Raj, and Amitabh Bachchan). Connections can be visibly traced to
DiCaprio (highlighted in green). So, for example, Donald Sutherland (near
the bottom) can be connected to DiCaprio via Robert De Niro. Amitabh
Bachchan and Christopher Lee (also near the bottom) are both two steps
away from DiCaprio and can be connected via Ben Kingsley.
