Using Punkt for Sentence Segmentation in non-Latin Scripts: Experiments on Kurdish (Sorani) Texts.
Roshna Omer AbdulrahmanHossein HassaniPublished in: CoRR (2020)
Keyphrases
- text segmentation
- segmentation algorithm
- image segmentation
- natural language
- medical images
- discourse structure
- segmentation method
- manually generated
- level set
- multiscale
- shape prior
- syntactic analysis
- fully automatic
- sentence level
- natural language generation
- energy function
- region growing
- segmentation accuracy
- semantic role labeling
- text corpus
- fully unsupervised
- machine learning