Does Social Media Big Data Make the World Smaller? An Exploratory Analysis of Keyword-Hashtag Networks

Social Media Big Data have transformed the scale of exploratory analysis on the web and offered new means of performing tasks that were not feasible before. Over a half century ago, Milgram showed that the average number of intermediaries (called the "degrees of separation") between two individuals is less than six (between 4.4 and 5.7). The study was done via means of regular mail on a very limited number of participants. With the advances of Internet, Dodds et al. performed the first electronic Milgram's using 60,000 participants and achieved similar results of Milgram's. These experiments remained within a very small-scale until the emergence of Facebook and social media. In 2012, Backstrom et al. carried out the largest Milgram-like experiment performed using the entire Facebook network. The findings showed that the average degrees of separation is 3.74, which suggest a shrinking world due to the impact of Big Data in a highly connected environment. Inspired by this recent finding, we pose the question: How does this newly observed phenomenon affect contents on Big Data environments? In this study we present an exploratory analysis of large-scale K-H networks generated from Twitter. We used two different measures (1) The number of vertices that connect any two keywords, (2) The eccentricity of keyword vertices, a well known path measure. Our analysis shows that K-H networks conform to the phenomenon of the shrinking world. Specifically, it shows that the number of vertices of any two keywords, that were not originally connected in the K-K networks, is exactly three while the eccentricity of every keyword is exactly four.