Tags, Graphs and Blog Posts

I last wrote about Impure.com and its network tools working with Google Webmaster Tools data.

Impure.com also has tools for dealing with RSS. These are meant to be used with blog feeds though, and are of limited use with other kinds of feeds, such as Blekko search RSS for example. However, it is possible to do some interesting things with blog posts and the tags that they have been assigned.

In this workspace I took four loosely related blogs, and combined their feeds into one bipartite graph. Each individual feed can also be explored separately.

Meaningful Visuals?

As well as being nice to look at, these graphs illustrate where a small set of blogs intersect in their chosen topics, within the constraints of the data sampled. There are a number of issues with relying heavily on tag data, but for broad categories, such as ‘SEO’, ‘Search’ and specific brand terms, it can be revealing.

The most useful data available in this exercise is linking all blog posts within the RSS feeds sampled to their tags. The ability to see:

• How close a blog’s posts group and
• How close separate blogs get

by their own subject material in a simple, easy to generate graphic is useful.

Limitations

There are some limits on this kind of analysis that mean it can’t replace more in-depth keyword or semantic tools and techniques. The RSS data has some limits, as tags on blog posts have some issues:

• Self selected by author
• Synonyms unaccounted for
• Variations in use
• Variations in frequency

And then what?

For examining the thematic relationship between a group of blogs, using Impure.com this way can still return value. Just for the ability to produce the visuals quickly and easily, it is worth looking at.