![]() I then cleaned and dumped these 8,603 tweets into a Google Sheet and shared it with my students: bit.ly/digisoc2020tweets. You might want to use datasets like Donald Trump’s tweets or State of the Union speeches. I used R’s “ rtweet” package and this Storybench tutorial to access the Twitter timelines of 11 Democratic candidates running in the 2020 election. This simple-to-use technique employs two simple formulae: =REGEXMATCH(C2, “keyword”) and =COUNTIF(J2,”TRUE”) to answer questions like “When and how do the candidates talk about race, guns or climate?” Getting the data I recently ran this tutorial in my journalism class at Northeastern University to help students answer some questions they generated around the Twitter timelines of the Democratic candidates running for U.S. ![]() While techniques for text mining, sentiment analysis and other natural language processing are ubiquitous on the Internet, they’re not always accessible to students.
0 Comments
Leave a Reply. |