Nepali News Corpus for Language Processing

A corpus of news articles in the Nepali language, hosted on Hugging Face by AdarshaShrestha. The dataset is categorized for natural language processing and text modality. It was last updated in March 2026.

Use Cases

• Train language models for Nepali text generation using the article text.
• Perform topic modeling or classification on news article categories.
• Develop named entity recognition systems for Nepali using entities within the news text.
• Analyze linguistic patterns and syntax specific to the Nepali language from the corpus.
• Benchmark machine translation models for English-to-Nepali or Nepali-to-English tasks using parallel news data if available.

Strengths

• Hosted on a major platform (Hugging Face) with an MIT license for broad use.
• Categorized for natural language processing with a text modality tag.
• Updated in 2026, indicating recent maintenance.

Limitations

• Unknown size, row count, and specific column structure.
• Potential geographic or topical bias limited to news sources available in Nepal.
• Lack of detailed documentation on collection methodology and article sources.

Provenance

Source

Hugging Face dataset uploaded by AdarshaShrestha.

Collection Method

Assembled as a corpus of Nepali news articles; specific collection method unknown.

Freshness

Last updated March 2026.

Geography

Presumably Nepal, based on language and content.