Downloads · 30 days
24
71% of all-time downloads
felipesuarezgiri/ESG_BiodiversityBERT_Commitment_Classification
ESG_BiodiversityBERT_Commitment_Classification is a machine learning model from felipesuarezgiri. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Biodiversity Commitment Classifier Model Overview: This binary text classification model identifies and distinguishes biodiversity commitments in corporate sustainability reports. It classifies paragraphs as either:
Downloads · 30 days
24
71% of all-time downloads
All-time downloads
34
Public
Parameters
82.3M
329 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors329 MB · 99%
From the Hugging Face model README
Biodiversity Commitment Classifier Model Overview: This binary text classification model identifies and distinguishes biodiversity commitments in corporate sustainability reports. It classifies paragraphs as either:
Commitment: The company pledges specific actions to improve biodiversity or reduce negative environmental impacts Non-commitment: General statements, observations, or non-actionable content related to biodiversity
Model Architecture: Built on ClimateBERT, a DistilRoBERTa-based model pre-trained on climate-related text, this classifier was fine-tuned specifically for biodiversity commitment detection in corporate disclosures. Training Data: The model was trained on a curated dataset of 2,000 manually annotated paragraphs extracted from sustainability reports of Fortune Global 500 companies, ensuring high-quality labels and real-world applicability to corporate ESG analysis.
Recommended Pipeline: First use ESGBERT/EnvironmentalBERT-biodiversity to identify biodiversity-related paragraphs, then apply this model to identify commitments.
Performance Metrics: Average of 5-fold cross-validation: Weighted F10.928 Weighted Precision0.930Weighted Recall0.929AUC-ROC0.976