tobischimanski's picture
Update README.md
a2678a6 verified
---
language: en
license: apache-2.0
datasets:
- ESGBERT/action_500
tags:
- ESG
- environmental
- action
---
# Model Card for EnvironmentalBERT-action
## Model Description
As an extension to [this paper](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4622514), this is the EnvironmentalBERT-action language model. A language model that is trained to better classify action texts in the ESG domain.
Using the [EnvironmentalBERT-base](https://huggingface.co/ESGBERT/EnvironmentalBERT-base) model as a starting point, the EnvironmentalBERT-action Language Model is additionally fine-trained on a dataset with 500 sentences to detect action text samples. The underlying dataset is comparatively small, so if you would like to contribute to it, feel free to reach out. :)
## How to Get Started With the Model
See these tutorials on Medium for a guide on [model usage](https://medium.com/@schimanski.tobi/analyzing-esg-with-ai-and-nlp-tutorial-1-report-analysis-towards-esg-risks-and-opportunities-8daa2695f6c5?source=friends_link&sk=423e30ac2f50ee4695d258c2c4d54aa5), [large-scale analysis](https://medium.com/@schimanski.tobi/analyzing-esg-with-ai-and-nlp-tutorial-2-large-scale-analyses-of-environmental-actions-0735cc8dc9c2?source=friends_link&sk=13a5aa1999fbb11e9eed4a0c26c40efa), and [fine-tuning](https://medium.com/@schimanski.tobi/analyzing-esg-with-ai-and-nlp-tutorial-3-fine-tune-your-own-models-e3692fc0b3c0?source=friends_link&sk=49dc9f00768e43242fc1a76aa0969c70).
It is highly recommended to first classify a sentence to be "environmental" or not with the [EnvironmentalBERT-environmental](https://huggingface.co/ESGBERT/EnvironmentalBERT-environmental) model before classifying whether it is "action" or not. This intersection allows us to build a targeted insight into whether the sentence displays an "environmental action".
You can use the model with a pipeline for text classification:
```python
from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline
tokenizer_name = "ESGBERT/EnvironmentalBERT-action"
model_name = "ESGBERT/EnvironmentalBERT-action"
model = AutoModelForSequenceClassification.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(tokenizer_name, max_len=512)
pipe = pipeline("text-classification", model=model, tokenizer=tokenizer) # set device=0 to use GPU
# See https://huggingface.co/docs/transformers/main_classes/pipelines#transformers.pipeline
print(pipe("We are actively working to reduce our CO2 emissions by planting trees in 25 countries.", padding=True, truncation=True))
```
## More details to the base models can be found in this paper
While this dataset does not originate from the paper, it is a extension of it and the base models are described in it.
```bibtex
@article{Schimanski23ESGBERT,
title={{Bridiging the Gap in ESG Measurement: Using NLP to Quantify Environmental, Social, and Governance Communication}},
author={Tobias Schimanski and Andrin Reding and Nico Reding and Julia Bingler and Mathias Kraus and Markus Leippold},
year={2023},
journal={Available on SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4622514},
}
```