Looking back on the analyses performed for #BlackLivesMatter, #NoDAPL, and #woke, the network analysis proved to be most informative. The network analysis revealed that #NoDAPL conversation clusters are centered primarily around news\/media sources. It was no surprise that the number of #NoDAPL tweets drastically decreased after the\u00a0protesters were removed by police. News\/media sources rarely report on social justice stories once the “drama” ends. The network analysis revealed that people who use #woke were not communicating with one another. Moreover, the topic of #woke tweets is not consistent and difficult to analyze. There was one user whose tweet was retweeted by many users The #woke student analyzer determined the user to be a drag queen, thus inferring that #woke is used by a variety of users for endless purposes. Finally, the network analysis revealed two main clusters of #BlackLivesMatter users. One cluster was around a white woman, and the other was around the official Black Lives Matter Twitter account. We expected that these two clusters existed because they used #BlackLivesMatter in different\/opposing ways. However, after determining the most retweeted tweets on Feb. 26, it was evident that everyone was tweeting about the death of Trayvon Martin.<\/p>\n
The six graphs\/ analyses varied in their level of helpfulness. As mentioned above, the network analysis proved to be most helpful for analyzing #BlackLivesMatter, #NoDAPL, and #woke as a group. Looking at just #BlackLivesMatter, the word cloud, word graph, and graph of tweets per day were also informative. The word cloud began to reveal that BLM supporters on Twitter are also trans rights supporters. This theory was confirmed by the word graph and graph of tweets per day. The Pie chart of language and the Map were less helpful. The pie chart confirmed the assumption that most tweets are in English, and the map failed to inform anything since there were so few mappable tweets.<\/p>\n
Presentation Outline:<\/p>\n