Component | Requires | Model | notes |
---|---|---|---|
ner_crf | sklearn-crfsuite | conditional random field | good for training custom entities |
ner_spacy | spaCy | averaged perceptron | provides pre-trained entities |
ner_duckling_http | running duckling | context-free grammar | provides pre-trained entities |
ner_mitie | MITIE | structured SVM | good for training custom entities |
NER中可以用正则表达式帮助CRF进行实体识别,在Training Data Format(https://rasa.com/docs/nlu/dataformat/)中,可以给出一个正则表达式列表,为ner_crf增加一个额外的特征项(1 or 0),表示是否检测到正则表达式。