liyi-cute 0.0.6

A text processing tools

liyi_cute: Text Tools

liyi_cute 是文本辅助工具，帮助NlPer减少模型输入前的预处理工作

online documentation:the read docs

Usage:

python setup.py install

pip install liyi_cute

install packages

pip install -r requirements.txt

数据结构

{
 "id": 1,
  "document": "xxxx",
  "": ""
}

信息抽取

实体抽取， 关系抽取，事件抽取， 属性抽取 以brat标注为例子: 标注文件开头标志 Entity: T

[entities]
Protein
Entity

T8	Negative_regulation 659 668	deficient
T9	Gene_expression 684 694	expression

{
"entities":[{"mention": "expression",
  "type": "Gene_expression",
  "start": 447,
  "end": 457,
  "id": "T1"}]
  }

Rlation: R

[relations]

Protein-Component	Arg1:Protein, Arg2:Entity
Subunit-Complex	Arg1:Protein, Arg2:Entity

R1	Protein-Component Arg1:T11 Arg2:T19
R2	Protein-Component Arg1:T11 Arg2:T18

## 暂时不支持
Equiv	Arg1:Protein, Arg2:Protein, <REL-TYPE>:symmetric-transitive
*	Equiv T3 T4

   {"relations": [{"type": "Part-of",
                 "arg1": {"mention": "c-Rel","type": "Protein","start": 139,"end": 144,"id": "T1"},
                 "arg2": {"mention": "NF-kappa B","type": "Complex", "start": 163, "end": 173, "id": "T2"},
                 "id": "R1"}]}

Event: E 暂时不支持

[events]

Gene_expression Theme:Protein
Binding Theme+:Protein

E3	Binding:T9 Theme:T4 Theme2:T5 Theme3:T6
E4	Binding:T20 Theme:T16 Theme2:T17 Theme3:T19

## 暂时不支持
E6	Negative_regulation:T10 Cause:E3 Theme:E5

Attribute: A 暂时不支持

[attributes]

Negation        Arg:<EVENT>
Confidence        Arg:<EVENT>, Value:Possible|Likely|Certain