Are topic models reliable or useful?
The final post in our series examines how topic models can and can’t help when classifying large amounts of text.
People in different subgroups within the U.S. population often have different views of, relationships with and priorities for the news media. Partisanship is among the strongest divides, with Republicans expressing much more negativity toward the media than Democrats (see Chapter 4). But striking divisions also emerge between other groups in the country, including racial and […]