Being able to visualize my manuscripts of
papers has always intrigued me I am a very visual person and I am therefore
naturally drawn towards visual aids when trying to understand thing.
The idea of visualizing a paper through the
links the references make has for some time been an idea I have mulled on. It
is only recently I have had the time to get to know the network visualizing
software Gephi and see if this tool could be used for the purpose. The results
you see below are the fruit of about 2-3 hours work. It is a small paper with
26 references and I have not included cross referencing between the individual
references, but this is definitely a next step.
There are some quite obvious observations
that can be made such as the largest bulk of the references appearing in the
introduction (node marked I) and the Discussion (node marked D). What is
surprising to me is that I have always thought I had a bigger bridge between
the Introduction and the Discussion. What the graph it shows is that I do not
reuse the references that I have introduced in the introduction. Whether this
is good I do not know but it is definitely a sign that I have tried to remove
redundant referencing by extending the information given in the introduction
rather than repeating it in the discussion.
Overall I am quite pleased with the result,
and it has given me the first ideas to what I need to do to perfect this
visualization.
1. I have to find a uniform
setting that will allow me to visualize my papers under the same constraints.
This will let me compare different papers directly.
2.
The links between the
individual references, although adding a sizeable amount of work gives an
important insight into whether the references used are redundant, fillers or
overused.
I am looking
forward to trying this on some of my larger papers. I might re post with a more
extended analysis of this paper.
