Sundanese Language Twitter Data

A collection of Twitter posts written in Sundanese, a low-resource language spoken in Indonesia. The dataset is sourced from the UCI Machine Learning Repository and is tagged for social media and natural language processing research. The specific temporal coverage, volume, and collection method are not detailed.

Use Cases

• Train language models for Sundanese using raw tweet text.
• Analyze code-switching patterns between Sundanese and Indonesian languages within tweets.
• Conduct sentiment analysis on Sundanese social media discourse.
• Study informal vocabulary and neologisms in a low-resource language from social media text.

Strengths

• Focuses on a low-resource language, providing data for an underrepresented NLP domain.
• Sourced from a real-world social media platform, reflecting contemporary informal language use.

Limitations

• Unknown dataset size, row count, and specific time range limit reproducibility.
• Lack of column descriptions or sample data prevents assessment of structure and annotation quality.
• Potential for geographic and demographic bias limited to Twitter users in West Java, Indonesia.

Provenance

Source

UCI Machine Learning Repository

Collection Method

Presumably scraped or collected from the Twitter platform.

Time Range

null

Freshness

null

Geography

Likely Indonesia, particularly West Java region where Sundanese is spoken.