Takes news transcripts scraped from archive.org and analyzes them for aggressive language
The data used in this project was gathered by Gaurav Sood, and can be found here
-
split_transcripts.Rmd cleans the data, splits each transcript into individual sentences, and outputs the result.
-
call_perspective_api.py assigns an aggression score to each sentence by calling the perspective API.
-
clean_local_call_nums.r takes a list of local channels and ties them to the corresponding national network.
-
chart_creation.r cleans the scored data and includes basic exploratory analysis.
local_channels.csv includes every local news channel in the country along with the corresponding national network. Data gathered here.