A dataset of the top 75 most variable log-transformed word counts for each US president aggregated over several speeches (Inaugural, State of the Union, etc.). Stop words have been removed and words have been stemmed.

presidential_speech

Format

A data.frame with 44 rows (one for each president) and 75 columns (log transformed word counts)

Source

http://www.presidency.ucsb.edu

Details

Grover Cleveland was elected president twice (1892 and 1884). For our purposes his speeches are combined.