You should understand text features, classification metrics, and feedforward neural networks. If you need those foundations, take Text Features and Neural Classification first. If you are new to NLP, begin with Introduction to Natural Language Processing, followed by Text Features and Neural Classification, before taking this course.

You’ll examine the limitations of word counts and TF-IDF, then explain how word embeddings address challenges in representing language. You’ll also explore how embeddings are learned and how pretrained embeddings can support text classification. In the mini project, you’ll justify an embedding strategy for classifying social media sentiment.

You’ll explore how recurrent neural networks process text sequences and why word order matters when working with language. You’ll use these concepts to justify architecture choices in a neural text classifier proposal, including the use of embedding, pooling, or recurrent layers.