Long-term word frequency dynamics derived from Twitter are corrupted: A bespoke approach to detecting and removing pathologies in ensembles of time series.
Peter Sheridan DoddsJoshua R. MinotMichael V. ArnoldThayer AlshaabiJane Lydia AdamsDavid Rushing DewhurstAndrew J. ReaganChristopher M. DanforthPublished in: CoRR (2020)