How Iranian Instagram Users Act for Parliament Election Campaign

Social media place where people communicate and share their ideas provide rich information for social network analysis. There are various analyses such as information diffusion modeling and community detection which are used to analyze data of social networks. In this paper, we investigate some novel aspects of hashtag diffusion among Iranian communities in Instagram in the period of the last legislative election in Iran. After data preparation, we analyze the validation of three different assumptions. First, we study the effects of follower-followee relations in the spread of the campaign hashtags. Based on the timestamps of the posts, we invoke NetRate method to estimate information diffusion rates over edges of follower-followee network. Then, by application of Louvain method as a community detection algorithm, we investigate the relation of community membership and contagion transmission rate. Finally, we study observed topical preferences in network communities. Results show the flow of information from followees to followers with a significant rate of diffusion over the whole network. However, being part of a specific community does not contribute to be exposed to a cascade faster than others. While the communities were defined based on modularity maximization and no information related to hashtags involved, a topical preference also is observed within the communities’ hashtags which had the same orientation as observed in two major political parties of Iran.