Skip to main content

Who Retweets Whom? A Quantitative and Qualitative Analysis of a Social Network

Social network analysis (SNA) is the process of investigating social structures using networks and graph theory. It characterizes networked structures in terms of nodes or vertex (individual actors, people, or things within the network) and the ties, edges, or links (relationships or interactions) that connect them. I won’t talk too much detail on the SNA theories as you can easily find them on the internet.

There are some sources that provide data for SNA such as Stanford and kdnuggets. In this simple analysis, I extract Twitter retweet data using “twitteR” package and analyse network retweet of a certain topic using “igraph” package in R. This tutorial from cosmopolitanvan may help you to replicate this work.

As one of the largest social networks on the Internet, Twitter can be used for expanding your business or website's audience. It is free to create an account, it is easy to start tweeting to promote your work or share your ideas and thoughts. Twitter has 284 million monthly users as of December 2014, according to Trusted Reviews.

It’s fun to analyse Twitter’s retweet data, we can see how users interact with each other retweeting a certain topic on a certain time, identify which user gets many retweets, and make a conclusion of the user’s preference e.g. political preference. It is also not surprising to see a lot of public figures retweet each other intensively.

The network graph below shows Retweet Network of #Jokowi, President of Indonesia, from 500 Twitter users (I use only 500 data since network analysis and visualization is computationally intensive). Size of yellow nodes can refer to indegree or outdegree centrality. In this case, it reflects indegree centrality or number of retweet a user gets. The biggest vertex or node belongs to @militanvespa which means this user gets the highest number of retweet on a certain time. We also can detect visually which user has high outdegree centrality or number of retweet a user does, the shape of the network looks like a “star”.


It’s interesting to see some public figures retweeting to each other, as what kind of idea/tweet regarding “jokowi” these users have shared or reposted is a different matter to discuss (we can check their tweets or guess based on the latest news). It is also important to describe the “specific time” in this analysis since it would not be relevant as the retweets may change in every minute or hour. We may also detect which account is suspected as a “buzzer” or social media influencer.


There is other useful statistics in “igraph” package which is betweenness centrality. Betweenness centrality is to measure how often a node lies on the shortest path between two other nodes. The more a node appears on one of those shortest paths, the higher its betweenness centrality. @militanvespa has the highest betweenness centrality, this user can be called a broker which means that information needs to pass through that entity to be shared by the other nodes. This also means that by cutting this node, chances are the network will fall apart into unconnected components.
Location: Melbourne VIC, Australia

Comments

Post a Comment

Popular posts from this blog

How to Create Indonesia Map in R

Creating the Map In this article, I will try to explain how to make Indonesia Map in R. I will assume that you are already familiar with the basic codes in R. First, we need the required libraries : require (maps) #loading maps package require (mapdata) #loading mapdata package library(ggplot2) #ggplot2 package library(readxl) #package for read .xlsx file library(ggthemes) #package for ggplot2 theme library(ggrepel) #extendig the plotting package ggplot2 for maps Then, we prepare the data that contains the information of provinces name, latitude, and longitude of every province in Indonesia, e.g. : You can download the data in here:  Data Now open the file and create the polygon: setwd( "your file's path" ) #set your own directory mydata<- read _xlsx( "dummy.xlsx" ) #assign the data to "mydata" View(mydata) #view the data, notice the column of "latitude","longitude", "woe_label" glo...
1 comment
Read more

What Can We Learn from Greek Debt Dramas?

Greek Debt Dramas Before the Global Financial Crisis (GFC) in 2008, the Greek had positive economic growth and it was considered high among countries in eurozone. Average economic growth reached almost four per cent between 1999 and 2007. Then the crisis hit in 2007 where housing bubble burst and made the subprime mortgage market in the United State collapsed. The crisis in the U.S. created a chain reaction which causing global banking crisis and credit crunch that lasts through 2009. The crisis made Lehman Brothers, big financial company, collapsed and the government in the United States and Europe prepared to bail out their banks. Greece failed to pay their huge debt since borrowing costs rose and financing dried up.  The financial crisis affected the Greek economy by reducing financial liquidity and business activity. Greece had been fortunate enough to face the crisis with the euro instead of its national currency, if they were using their national currency the crisis wo...
Post a Comment
Read more

Empirical Evidence of Engel’s Law Among Social Grant Recievers

Engel's law is an observation in economics stating that as income increases, the proportion of income spent on food decreases, even if absolute expenditure on food increases. The law was named after the statistician  Ernst Engel (1821–1896). One application of this statistic is treating it as a reflection of the living standard of a country. As this proportion — or "Engel coefficient" — increases, the country is by nature poorer; conversely a low Engel coefficient indicates a higher standard of living. Engel's Law image source: Wikipedia Using data collected through National Social and Economic Survey (NSES) by BPS-Statistics Indonesia, I tried to examine the existence of Engel's Law among households that received social grants in West Papua-Indonesia. Some studies found that giving additional money to the low-income households resulted in an increase in overall expenditure on food (on absolute) but the proportion  of income spent on food would decrea...
Post a Comment
Read more