Skip to main content


This is a project from a recent Cornell graduate that embodies the idea of analyzing connections within a social network. It is essentially a Facebook app (whatever it may be called) that analyzes your network of friends and gives you an output of the various friend groups, or clusters as they put it, you have and are a part of. The idea is that when inviting friends to use a particular app or service on Facebook, you are more likely to invite a close friend rather than a mere acquaintance. The current system has your friend list organized alphabetically but your close friends could easily be far lower on the alphabetical list meaning the chances of you recommending the page/app/feature is lower. So in short, this Facebook app analyzes your friends and shows the various networks within your friend network which has more practical purposes than expected.

As discussed earlier and in class, how close you are to a friend influences the amount of communication with them and willingness to make recommendations. This is essentially seeing whether a link between two individuals is strong or weak. This app does even more than that and gauges the strength of the connections within a cluster and the strength of the connections between the individuals in the cluster and the individuals outside the cluster. This takes the concept that we discussed in class who would be better to suggest friends to other friends (in this case it could be anything), person A or person B. Strength of friendship was determined to be pretty important of a factor. This program allows to determine who would be most likely to make suggestions and suggestions that matter. This app uses existing friend links and mutual friends to figure out the clusters and strength of the connections amongst other tools. Using networks to analyze networks and putting it to very practical use.



One Response to “ GraphMuse ”

Leave a Reply

Blogging Calendar

September 2012
« Aug   Oct »