THUCnews: Classification Performance Metrics on a 10% Sample

Classification performance on a 10% subset of the THUCnews Chinese news text dataset. The 5.5 KB XLS file was authored by Yumin Yan and last updated on May 27, 2026. Its specific metrics likely include accuracy, precision, recall, or F1 scores for text categorization tasks.

Use Cases

• Benchmarking model performance on Chinese text classification based on the described evaluation metrics.
• Comparing algorithm accuracy on a subset of the THUCnews corpus.
• Validating text classification pipelines using pre-computed performance scores.

Strengths

• File size is precisely 5.5 KB, indicating a compact and easily shareable dataset.
• License is clearly defined as CC-BY-4.0, facilitating open reuse.
• Last update timestamp is provided (2026-05-27 17:28:33).

Limitations

• Row count is unknown, which may limit suitability assessment for statistical analysis.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset represents only a 10% sample, which may not reflect performance on the full corpus.

Provenance

Source

Yumin Yan via figshare

Collection Method

Likely derived from evaluating models on the THUCnews dataset.

Freshness

Last updated 2026-05-27 17:28:33; freshness should be verified.