Abstract:
Online social networks (OSNs) provide a platform that allows actors to disseminate information in the form of user-generated contents as well as receive updates from their contacts. Organizational users employ OSNs to diffuse product information, technologies, knowledge and innovation. In general, OSN can be classified as textual based OSNs and visual based OSNs. Considerable research has been invested in examining user behaviours, network structures, and patterns of evolution for various OSNs, such as, Facebook, Flickr, Okurt, Twitter, Myspace, Live journal and YouTube. This research explores information diffusion issues in OSNs. This study becomes part of the social network research paradigm, but with fresh objectives, namely: (1) identifying the information diffusion patterns, (2) distinguishing between textual based online social networks and visual based online social networks, (3) examining the differences of diffusion patterns among three selected accounts, and (4) addressing the phenomena of evolution for each account with the implementation of trend analysis. Raw data for this research are collected from three accounts - BBC World News, Pure New Zealand and WWF Climate using both Flickr (a visually orientated OSN) and Twitter (a textually orientated OSN). NodeXL is adopted in this thesis for data collection and metrics generation. In order to examine information diffusion structurally, actors in the egocentric networks are classified into different ranks on the basis of degree centrality. Findings of this thesis are obtained through interpreting key metrics which include degree centrality, eigenvector centrality, clustering coefficient, diameter and average geodesic distance for both core actor (the account holder) and first rank actors. Notably, reciprocity in this thesis is identified as the key metric to examine information diffusion for OSNs. Both dataset descriptions and statistical tests suggest that information diffusion in Flickr is more efficient than in Twitter due to the existence of small diameter and average geodesic distance. Twitter is advantageous for communications intensity, frequency and the volume of transmitted information. Results of account comparisons indicate that reciprocity is positively correlated with the improvements in account populations and information diffusion.