language: | |
- zh | |
license: apache-2.0 | |
tags: | |
- bert | |
- ner | |
- resume | |
pipeline_tag: other | |
widget: | |
- text: "欧阳可守,2000年出生,硕士学历毕业于清华大学,目标职位:自然语言算法工程师," | |
example_title: "Resume Entity Recognition" | |
## 简历命名实体识别 | |
- 使用 | |
```python | |
ner_predictor = pipeline( | |
task="nerpipe", | |
model="minskiter/resume-token-classification-extends-0708", | |
device="cpu", | |
trust_remote_code=True | |
) | |
``` | |
- 模型 | |
BertCrfModel | |
- 数据集 | |
Resume | |
Typing| Train | Dev |Test | |
:-:|:-|:-:|:-: | |
Sentence | 3.8k | 0.46k | 0.48k | |
Char | 124.1k | 13.9k | 15.1k | |
Tag | Meaning | Train | Dev |Test | |
:-:|:-|:-:|:-:|:-: | |
CONT | Country | 260 | 33 | 28 | |
EDU | Educational Institution | 858 | 106 | 112 | |
LOC | Location | 47 | 2 | 6 | |
NAME | Personal Name | 952 | 110 | 112 | |
ORG | Organization | 4611| 523 | 553 | |
PRO | Profession | 287 | 18 | 33 | |
RACE | Ethnicity Background | 115 | 15 | 14 | |
TITLE| Job Title | 6308| 690 | 772 | |
Total Entity |--- |13438| 1497| 1630 |