Chronotope’s avatarChronotope’s Twitter Archive—№ 91,452

  1. …in reply to @profcarroll
    profcarroll I think it depends on how the data is collected. If it's aligning targeting terminology w/result of in-house user studies--as opposed to mass user data collection--which seems likely in a news org without access to the usual sentiment indicators, then it's somewhat different.
    1. …in reply to @Chronotope
      profcarroll ESPN's approach -- which requires large swaths of data collection -- is not great. But I'm not sure that's NYT's approach?


Search tweets' text