NUSS: An R package for mixed N-grams and unigram sequence segmentation

NUSS is an R package for segmenting compound text sequences, such as hashtags, using a dual strategy that integrates n-gram segmentation with unigram sequence segmentation. Designed for context-sensitive analysis without training data, it builds custom dictionaries from any corpus and balances segmentation accuracy with computational efficiency. NUSS enhances preprocessing pipelines in text mining, particularly for social media data, and enables robust analysis in sentiment research, topic modeling, and digital discourse studies. It is the first CRAN-published R package dedicated to text segmentation.