Deep Learning for Text Analytics

The module Deep Learning for Text Analytics (DELTA) covers contemporary practices for natural language processing using deep neural networks. 

The module is designed as a follow-up course to our Business Analytics and Data Science course. It targets students who seek to further develop their understanding of and skills in applied machine learning (ML), especially in the direction of leveraging textual data. We assume that students taking this module are familiar with standard ML workflows and algorithms, and have solid skills in Python programming. 

Topics covered in DELTA include but are not limited to:

• Fundamentals of artificial neural networks
• Neural network architectures for sequential and unstructured data
  • Recurrent and gated neural networks
  • Convolutional neural networks
• Fundamentals of natural language processing (NLP)
  • Learning word embeddings using word2vec and cousins
  • Language modeling
• State-of-the-art NLP approaches
  
  • Attention and Transformers
  • NLP transfer learning

Although a key focus of the module is to elaborate on the functioning of deep learning techniques for text as well as other types of data, it is a given that we also elaborate on (business) applications of the corresponding methodology. The computer exercises use the Python programming language to familiarize students with contemporary libraries for deep learning (e.g., Keras) and NLP (e.g., NLTK)