Improving Chinese Segmentation-free Word Embedding With Unsupervised Association Measure.
Yifan ZhangMaohua WangYongjian HuangQianrong GuPublished in: CoRR (2020)
Keyphrases
- word segmentation
- chinese word segmentation
- pos tagging
- pointwise mutual information
- word recognition
- chinese text
- chinese text retrieval
- image segmentation
- similarity measure
- text segmentation
- n gram
- unknown words
- fully unsupervised
- bayesian image segmentation
- word level
- segmentation algorithm
- level set
- numeral strings
- syntactic categories
- co occurrence
- segmentation method
- error measure
- language independent
- unsupervised learning
- cross lingual
- shape prior
- region growing
- edge detection
- vector space
- distance measure
- keywords
- english text
- text classification
- handwritten documents
- related words
- watermarking algorithm
- object segmentation
- information theoretic
- image regions
- energy function