The course provides in depth knowledge on the essential elements of Natural Language Processing (NLP), particularly based on machine learning and neural networks. The lectures cover topics such as text processing, language modeling, word embeddings, and sequence embeddings, and studies the application of these to document classification, sentiment analysis, information retrieval, computational social science, and detection of societal biases.

Covered topics in lectures:

• Text processing
• Sentiment analysis with machine learning
• Language modeling
• Count-based and neural word embedding models (word2vec, GloVe, etc.)
• Introduction to large language models (BERT) 
• Neural Information Retrieval
• Footprint of societal phenomena and biases in NLP

Information for the current semester (if available):

Prerequisites:

• For the lectures (VL), knowing about machine learning and neural networks methods is suggested, while the course also briefly covers these topics.
• For the practical part (UE), good Python programming skill is required.

Teaching materials of the latest course (Winter Semester 2022/23):

• Text Precoessing  slides, opens an external URL in a new window
• Document Classification - BoW and Semantic Methods slides, opens an external URL in a new window
• Word Embedding with Matrix Factorization slides, opens an external URL in a new window
• Neural Networks for NLP – a Walkthrough slides, opens an external URL in a new window
• N-Gram Language Models slides, opens an external URL in a new window
• Neural Word and Sentence Embeddings slides, opens an external URL in a new window
• Introduction to Large Language Models slides, opens an external URL in a new window
• Information Retrieval - Principles and Recent Approaches slides, opens an external URL in a new window
• Fairness and Societal Biases in NLP slides, opens an external URL in a new window

Follow-on course: Special Topics: Natural Language Processing with Deep Learning (KV)