---
language: de
tags:
- text-classification
license: "other"
metrics:
- f1
- precision
- recall
library_name: spacy
---

# Extraction of Formal Educational Requirements from Online Job Advertisements

## Usage

```python
from huggingface_hub import snapshot_download
import sys

# download snapshot of model
path = snapshot_download(
    cache_dir="tmp/",
    repo_id="bertelsmannstift/oja_education_extraction",
    revision="main",
    token=HF_TOKEN,
)

# Add pipeline module to path and import 
sys.path.append(path)
from pipeline import PipelineWrapper

# Init model
pipeline = PipelineWrapper(path=path)

# Predictions
queries = [{"posting_id": "123",
         "full_text": "Wir sind Firma XYZ. Wir suchen einen Data Scientist. Sie haben Mathematik, Politikwissenschaften oder ein vergleichbares Fach studiert.",
         "candidate_description": None,
         "job_description": None}]

result = pipeline(queries)


```


## Input:

```
queries = [
    {
        "posting_id": "foo",
        "full_text": "bla bla",
        "candidate_description": "bla bla",
        "job_description": "bla bla",
    },
    ...
]
```

## Output: 
```
[{'posting_id': 'foo', 
'education_level_raw_id': [...], 
'education_level_isced_id': [...], 
'education_studies_label': [...], 
'education_vocational_label': [...]},
...
]
```

## Variables and Taxonomy

- education_level_raw_id:
	- A code for a granular kind of education level in accordance with the German educational system. 
	- Classes: TODO
	- Rules for pattern matcher: https://huggingface.co/bertelsmannstift/oja_education_extraction/blob/main/assets/patterns_education_level.json
- education_level_isced_id:
	- Assignment of the detected granular education levels to international ISCED taxonomy (first two numbers)
	- Modifications: TODO
	- Classes: TODO
	- Assignment rules: https://huggingface.co/bertelsmannstift/oja_education_extraction/blob/main/assets/map_education_level_isced.json
- education_studies_label:
	- A list of academic subjects if asked for
	- Taxonomy follows DESTATIS Fächersystematik 2023: https://www.destatis.de/DE/Methoden/Klassifikationen/Bildung/studenten-pruefungsstatistik.pdf?__blob=publicationFile
	- Examples for few-shot training: https://huggingface.co/bertelsmannstift/oja_education_extraction/blob/main/models/studies/sbert_entity_classifier/examples.json
- education_vocational_label:
	- A list of vocational subjects if asked for
	- Based on official list of German vocational subjects by Bundesagentur für Arbeit. Can be obtained here: https://download-portal.arbeitsagentur.de/
	- Additionally, subjects have been grouped into 20 Clusters
	- Subjects with different specializations have been grouped as well
	- One additional supercluster: "Handwerk"
	- Examples for few-shot training: https://huggingface.co/bertelsmannstift/oja_education_extraction/blob/main/models/vocational/sbert_entity_classifier/examples.json

## Performance and Datasets


| Task                  | n Data | Sampling                                             | Annotators | Annotator Overlap | Inter-Annotator Agreement         | Test Split | Evaluation Mode           | Precision (Micro) | Recall (Micro) |
| --------------------- | ------ | ---------------------------------------------------- | ---------- | ----------------- | --------------------------------- | ---------- | ------------------------- | ----------------- | -------------- |
| Localization          | 1500   | Stratified by first number ISCO, equally distributed | 4          | 10 %              |                                   | 0.2        | NER partial               | 0.95              | 0.92           |
| Education Level (raw) | 1200   | Stratified by first number ISCO, equally distributed | 3          | 20 %              | 0.84 (Krippendorff)               | 0.5        | Multilabel classification | 0.92              | 0.87           |
| Academic Subject      | 1200   | Stratified by first number ISCO, optimized*          | 5          | 20 %              | 0.77 (Krippendorff)               | 0.2        | NER strict                | 0.87              | 0.88           |
| Vocational Subject    | 1200   | Stratified by first number ISCO, optimized*          | 2          | 20 %              | 0.72 (Krippendorff)               | 0.2        | NER strict                | 0.86              | 0.83           |

Note: since tasks are in pipeline, errors might accumulate