metadata
dataset_info:
features:
- name: noteId
dtype: int64
- name: noteAuthorParticipantId
dtype: string
- name: createdAtMillis
dtype: int64
- name: tweetId
dtype: int64
- name: classification
dtype: string
- name: believable
dtype: string
- name: harmful
dtype: string
- name: validationDifficulty
dtype: string
- name: misleadingOther
dtype: int64
- name: misleadingFactualError
dtype: int64
- name: misleadingManipulatedMedia
dtype: int64
- name: misleadingOutdatedInformation
dtype: int64
- name: misleadingMissingImportantContext
dtype: int64
- name: misleadingUnverifiedClaimAsFact
dtype: int64
- name: misleadingSatire
dtype: int64
- name: notMisleadingOther
dtype: int64
- name: notMisleadingFactuallyCorrect
dtype: int64
- name: notMisleadingOutdatedButNotWhenWritten
dtype: int64
- name: notMisleadingClearlySatire
dtype: int64
- name: notMisleadingPersonalOpinion
dtype: int64
- name: trustworthySources
dtype: int64
- name: summary
dtype: string
- name: isMediaNote
dtype: int64
- name: climate_label
dtype: int64
- name: climate_score
dtype: float64
splits:
- name: train
num_bytes: 5823325
num_examples: 10000
download_size: 2377349
dataset_size: 5823325
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
This dataset was created by running climatebert/distilroberta-base-climate-detector on the "summary" column of the first 10,000 rows of the Twitter community notes dataset (description, download link), to identify which notes relate to climate change.