Chinese word segmentation with minimal linguistic knowledge: An improved conditional random fields coupled with character clustering and automatically discovered template matching.
Richard Tzong-Han TsaiHong-Jie DaiHsieh-Chuan HungCheng-Lung SungMin-Yuh DayWen-Lian HsuPublished in: IRI (2006)
Keyphrases
- template matching
- conditional random fields
- linguistic knowledge
- natural language processing
- information extraction
- graphical models
- probabilistic model
- hidden markov models
- markov random field
- natural language
- higher order
- clustering algorithm
- matching algorithm
- generative model
- object recognition
- maximum entropy
- named entity recognition
- pairwise
- word sense disambiguation
- k means
- image matching
- dialogue system
- machine learning
- part of speech