| Popis: |
Since the US plays a crucial role in global politics, the presidential elections have been closely monitored. One of the ways to inspect the elections is through social media. Twitter is among the most popular social media used to examine public opinion because of the availability of datasets. It is interesting to explore and compare what the US netizens say about the Republican and Democratic presidential nominees for president in the 2020 US election (second debate) on Twitter, as well as the characteristics of users based on their ID names. The dataset used in this study is derived from Tweetsets. The dataset is analyzed under the framework of Social Network Analysis and Discourse Analysis. TF-IDF scores will be employed to extract unique words in each community in the networks. The results show that the relationship of tweeters in @realDonaldTrump network is more extensive, and the network has more active tweeters. According to text analysis, the content in @realDonaldTrump network contains more criticisms and sarcastic messages. The tweets in the network are also more diverse in terms of people who are mentioned, though the usernames in the network are less meaningful and interpretable. @JoeBiden network has almost two times less nodes but is more dense. Although tweet texts in the network also contain criticisms, there are more positive messages as compliments. |
