I came across this really nifty R packag called RLinkedIn and thought I might use it to quickly look through my own LinkedIn profile, connections and groups to see what I could find. There is a nice introduction on how to get the authorization done and get started with RLinkedIn here. I used that as a starting point to analyze my profile. Here are the results!
I started out by trying to see the industry-wise break-up of my connections. I decided to use a wordcloud to visualize the results. There really are no surprises there since the big ones, viz., information technology, research, services, software, human, resources, management all deal with aspects of my employment history. It is still interesting to see some of the other industries represented here.
I then decided to take a look at the countries where my connections are located. And of course, there is only one visualization that suggests to em when I have to look at country-specific data and that is on a map. So the map you see below is colored by the number of people from each of those countries who are my connections. Again, India dominates (the country I am from and am a current resident of), along with the US (the country I did grad school in). East Asia, where I spent almost 8 years is also fairly well represented. The two countries in Africa surprised me.
I then proceeded to look at the number of connections, that each of my connections had. You can see the histogram of the same below. From a network theoretic perspective, it is good to be connected to people who themselves have a number of connections and as you can see, more than half my connections seem to have > 500 connections themselves.
I finally decided to take a look at what people were talking about in the groups that I am a member of. My favorite visualization for representing word-related data is the wordcloud (if you haven’t figured it out already!). The data is restricted to the last ten posts in each group. I looked at all the words from the post summaries. Some of the words I would expect to be well represented are there including data, research, analytics, management etc. Corrosion was a big surprise for me. Either the group is really really active (and I need to see which group it is) or else someone is just super interested in corrosion 🙂 Of course I have a lot of friends who are deep into corrosion or other aspects of electrochemistry and it could just be a result of that.
Anyway, I thought this was an interesting way to look at one’s own LinkedIn profile. There is much more that one can do with the package include searching for jobs, companies, people as well posting messages onto specific groups. If you want the R code, do drop me a message with your email address and I shall email it to you.