Update README.md

README.md

@@ -1,3 +1,6 @@
 ---
 license: apache-2.0
+language:
+- de
 ---
+"# German FinBERT (Further Pre-trained Version)\n\nThis model card details the further pre-trained version of German FinBERT, a language model focusing on the financial domain within the German language.\n\n## Overview  \n**Author:** Moritz Scherrmann【23†source】  \n**License:** *Information not provided in the document*  \n**Framework:** BERT-base  \n**Language:** German  \n**Specialization:** Financial textual data  \n**Original Model:** gbert of deepset  \n\n## Pre-training Corpus\nThe pre-training corpus consists of German financial textual data. It comprises a comprehensive collection that includes financial reports, ad-hoc announcements, and news related to German companies. The corpus size is on par with those used for standard BERT models, indicating substantial coverage and depth.\n\n## Performance  \n### Fine-tune Datasets\nGerman FinBERT has been evaluated on three finance-specific tasks against generic German language models, showing improved performance in:\n- Sentiment prediction\n- Topic recognition\n- Question answering\n\nThe model effectively captures domain-specific nuances, outperforming standard models on finance-related texts.\n\n### Benchmark Results\n- *The precise benchmark results and comparisons are not provided in the accessed part of the document.*\n\n## Authors  \n**Moritz Scherrmann**  \nInstitute for Finance & Banking  \nLudwig-Maximilians-Universität München  \nLudwigstr. 28, RB 80539 Munich, Germany  \nEmail: scherrmann@lmu.de【23†source】\n\nFor additional details regarding the performance on fine-tune datasets and benchmark results, please refer to the full documentation provided in the study. German FinBERT represents an innovative development in the field of financial NLP, offering enhanced capabilities for analyzing German financial texts."