Natural Language Processing

Summary

This course provides an overview of natural language processing (NLP) on modern Intel® architecture. Topics include:

  • How to manipulate text for language models
  • Text generation and topic modeling
  • The basics of machine learning through more advanced concepts

By the end of this course, students will have practical knowledge of:

  • Application of string preprocessing techniques
  • How to apply machine learning algorithms for text classification and other language tasks

The course is structured around eight weeks of lectures and exercises. Each week requires three hours to complete.

Prerequisites

Python* programming

Calculus

Linear algebra

Week 1

This class introduces the uses and history of NLP. Topics include: 

  • The history of NLP and how it is used in the industry today
  • How to parse strings using powerful regular expression tools in Python
Week 2

This class teaches how to use NLP toolkits and preprocessing techniques. Topics include:

  • Explore techniques such as tokenization, stop-word removal, and punctuation manipulation
  • Implement such techniques using Python libraries such as NLTK, TextBlob, spaCy, and Gensim
Week 3

This class introduces how to measure similarity between words. Learn more about:

  • Levenshtein distance, which is used to compare the similarity of two words
  • How computers encode pieces of text into a document-term matrix and what the bag of words assumption is
Week 4

This class shows how machine learning is used for basic text classification. Topics include:

  • The basics of machine learning and a refresher on the terminology
  • A typical machine learning workflow for two different machine learning approaches to classify emails as either spam or not spam
Week 5

This class teaches an algorithm for natural language understanding and topic modeling. Learn more about:

  • How to use the latent Dirichlet allocation algorithm to extract topics from the document-term matrices
Week 6

This class continues to teach how to model and extract topics in text. Learn more about:

  • Alternative algorithms for discovering the topics embedded in texts
Week 7

This week teaches machine learning algorithms for NLP. Topics include:

  • How to use a neural network to transform words into vectors
  • Potential applications of these vectors (such as text classification and information retrieval)
Week 8

Continuing with the topic of machine learning, this class teaches more about applying neural networks. Topics include:

  • Text generation using Markov chains and recurrent neural networks
  • Advanced topics in NLP, such as seq2seq