Sign in to view source links and access this dataset
Description
552,821 unique text samples with 570,457 emotion label assignments form this large-scale, multilabel dataset. It aggregates data from six prominent emotion datasets: MELD, GoEmotions, TwitterEmotion, ISEAR, SemEval, and Crowdflower. The dataset was created by 'cirimus' and was last updated on July 28, 2025.
Use Cases
Train multilabel emotion classifiers based on the aggregated text samples from six source datasets.
Benchmark model performance on emotion recognition tasks using the large-scale, aggregated label assignments.
Study cross-dataset emotion representation and label consistency across the six integrated sources.
Strengths
Large scale with 552,821 unique text samples.
Aggregates data from six established emotion datasets, potentially increasing diversity.
Provides 570,457 total emotion label assignments, indicating multilabel annotations.
Limitations
Column-level documentation is absent; field semantics must be inferred after download.
Row count is unknown, which may limit suitability assessment.
Data may reflect source bias inherent to the six aggregated datasets.
Provenance
Source
Aggregated from six datasets: MELD, GoEmotions, TwitterEmotion, ISEAR, SemEval, Crowdflower.
Collection Method
Aggregation from existing datasets; full construction process documented in an accompanying paper.
Freshness
Last updated 2025-07-28 04:28:27; freshness should be verified.
License is unknown; terms of use must be verified before application.