MSIA 414: Text Analytics

Quarter Offered

Fall ; Yuri Balasanov


The course explores a breadth of Natural Language Processing (NLP) applications with a focus on contemporary, state-of-the-art systems, often based on deep learning techniques. Topics include word embeddings and common deep learning NLP architectures; approaches to a variety of NLP tasks such as text classification, named entity recognition, machine translation, information retrieval, etc. An independent project offers an in-depth exploration of an NLP topic of choice, including a review of relevant academic literature, machine learning experiments, system development and “productization”.  The course also includes a variety of practical industry NLP topics such as industry datasets, machine learning engineering, tracking and reproducibility of experiments, frameworks and deployment consideration, model explainability, and model “productization”.

Course objectives:

  • Develop familiarity with a variety of NLP applications and state-of-the-art solutions
  • Develop skills to understand non-trivial scientific NLP publications and NLP / ML libraries and framework for continuous and independent learning
  • Develop NLP / ML engineering skills and familiarity with common industry NLP tasks
  • Develop skills to creatively approach business problems and create practical NLP solutions