• Text representation by word frequency models (Bag of Words, Term Frequency Inverse Document Frequency models)
• Text analysis by vector embeddings: Word2Vec, Doc2Vec
• Neural networks
• Self-Attention of the transformer model
• Encoder and decoder language models
• Pretraining and finetuning of encoder models
• Application of text models in the financial domain

Students who participate in this course:
• develop a deep understanding of how modern text models work
• establish the connection between general methods of machine learning and modern text modeling
• evaluate which form of text analysis is suitable for different subject matters
• use state-of-the-art text models for the analysis and evaluation of important documents from the field of economics and business studies