How indonlp/nusax works
NusaX is a multilingual language-resource release: users access parallel machine-translation data, sentiment-analysis data, and sentiment lexicons for Indonesian languages. The README describes the sentiment datasets as translations verified by native speakers. The repository also includes a classical sentiment-analysis notebook, but its contents were not sampled, so no connection to particular datasets is asserted.
Main entry points: mt Translation corpus · sentiment Sentiment datasets · lexicon Sentiment lexicons
Components 5
Language datasets
- Translation corpus
datasets/mt - Sentiment datasets
datasets/sentiment - Sentiment lexicons
datasets/lexicon
Analysis code
- Classical analysis
code/notebook/sentiment_analysis_nusax_classical.ipynb
Other components
- Dataset user
Connections 3
- Dataset user → Translation corpus: accesses
README.md - Dataset user → Sentiment datasets: accesses
README.md - Dataset user → Sentiment lexicons: accesses
README.md