Clusterschmuck..

I thought I’d written about this kind of thing before, but having a quick search I don’t think that I have.  It’s a geeky marriage of the kind of thing I do at work (on occasions, although less so these days) and my personal life thanks to the Facebook Report over at Wolfram Alpha.

Essentially, taking a whole load of data (ie, your Facebook stuff) and doing funky things with it.  My favourite is this clustering exercise of friends based on connections with one another.  Certainly its’ first attempt gives some fairly neat clusters (I added the shapes and labels on there).  I just refreshed the page and noted that it comes up with a different map, although I’d still identify similar clusters from it.

clusteringOf course, as with any large dataset there is a bit of a problem when trying to shoehorn such a large group of people (504 were included in this analysis apparently) into ten buckets.  I’ve got outliers in there who don’t really fit in any of them, and plenty of ‘node’ people who traverse many of my different worlds (the ‘Family’ group in this instance holds many of these folk).

The categorisations that it tries to apply take workplaces/places of education and geographic information to try to inform the clusters (hence the colour coding).  Interestingly it doesn’t pick out ‘Likes’ or interests that might give some clues – but usually the linkages between people are enough to spot how these people relate to one another in your life.

The report goes further to help you identify these (although you’d know them intuitively I think), Rich my brother shares 149 friends on Facebook with me, Rich Crouch is next in line on the mutual friend count – crossing over as he does on the Forest, Ferocious Dog, Boots and probably sneaks well into the Family group of friends too.  It’s really interesting to see such things quantified.

The demographic splits are interesting too – two thirds of my Facebook friends are male, a third female, I’ve discovered that one of my old school friends is apparently 89 years old and therefore my oldest friend.  A timely lesson in the quality of data impacting the quality of results I guess!  The distribution of ages of my friends unsurprisingly centres closely around my own age range.

Geographically I’m very United Kingdom biased but do have a few further flung friends around the globe too!

Screen Shot 2013-10-13 at 20.52.31As well as stalking my friends, the report also looks at the content I bombard Facebook with.  Apparently I write DAY and TODAY a lot, as well as clearly mentioning our feathered critters by name quite a bit, and there’s more than hint in there that I might talk about Ferocious Dog rather a lot!  Partly given away by Ferocious, but also the easily misunderstood reference to dogging that makes it in there too!

Ultimately very pointless, although I suppose illustrative of the powerful information Facebook has to better understand people.  From a professional perspective the ability to analyse how people connect together is fascinating and something missing from our individualistic datasets at the moment.

Information is powerful stuff innit, whilst I’ve found this exercise pretty insightful it’s also maybe fuel to the fire that choose to ignore Facebook through fear of giving away too much information about themselves to cynical corporations.

If you’re not one of those, you can generate exciting maps and more from your own Facebook data by clicking on the link included up there somewhere!

 

Advertisements
Categories: blog | Leave a comment

Post navigation

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Create a free website or blog at WordPress.com.

%d bloggers like this: