Archeobertje
ArcheoBERTje is a Dutch BERT model tailored for the field of archaeology, fine-tuned from the BERTje model specifically for processing Dutch archaeological excavation reports.
Downloads 18
Release Time : 3/2/2022
Model Overview
This model is based on the Dutch BERTje model and fine-tuned using a corpus of approximately 60,000 Dutch excavation reports (around 650 million tokens), making it suitable for natural language processing tasks in archaeology.
Model Features
Specialized for archaeology
Optimized for Dutch texts in the field of archaeology, enabling better understanding and processing of professional terminology and context in excavation reports.
Fine-tuned from BERTje
Fine-tuned from the Dutch BERTje model, inheriting its excellent Dutch language processing capabilities while further adapting to the needs of archaeology.
Large-scale training data
Trained using a corpus of approximately 60,000 Dutch excavation reports (around 650 million tokens), covering a wide range of archaeological content.
Model Capabilities
Text classification
Named entity recognition
Text generation
Semantic understanding
Use Cases
Archaeological research
Automatic classification of excavation reports
Automatically classify archaeological excavation reports to facilitate quick retrieval and analysis by researchers.
Archaeological entity recognition
Identify archaeological terms, place names, artifact names, and other entities in reports to assist research analysis.
Cultural heritage management
Cultural heritage document processing
Automate the processing of cultural heritage-related documents to improve management efficiency.
Featured Recommended AI Models