Large Movie Review Dataset for Sentiment Classification

Serving as part of the Massive Text Embedding Benchmark (MTEB) for evaluating text embedding models on sentiment classification tasks. It contains movie reviews categorized for sentiment analysis, with a size category indicating between 10,000 and 100,000 entries.

Use Cases

• Evaluate embedding models on sentiment classification tasks using the MTEB framework.
• Train sentiment analysis models on monolingual English movie review text.
• Benchmark model performance on the specific MTEB task ID for sentiment scoring.

Strengths

• Part of the standardized Massive Text Embedding Benchmark (MTEB) for model evaluation.
• Dataset size is categorized as between 10,000 and 100,000 entries.
• Text modality is monolingual English, providing a focused language domain.

Limitations

• Exact row count, column structure, and sample data are unavailable for detailed inspection.
• The dataset's specific creation method and annotation details are not provided.
• Limited to the domain of movie reviews, which may not generalize to other text types.

Provenance

Source

Derived from the Large Movie Review Dataset, referenced in ACL anthology paper P11-1015.

Collection Method

Annotations are derived, but the specific gathering process is not detailed.

Freshness

Dataset metadata was last updated on 2025-07-06.

Geography

Region is indicated as 'us'.