Through this tutorial, you’ll get an introductory overview of how to make sense of a large corpus of textual data through SA. You will take advantage of open-source Python libraries (notably the Natural Language Toolkit and spaCy) to conceptualize, to design, and to implement distinct components of an SA model and workflow. After building an SA model from scratch, you’ll apply it to extract sentiments from customer review data. You’ll also get a chance to work with a pre-trained model on a corpus more representative of actual customer reviews (because they constitute an invaluable resource allowing organizations to understand quantitatively and to improve services, products, customer experiences, workforce analytics, social media monitoring processes, and, ultimately, business performance).

Approximate outline: (10 min) Setup: We’ll ensure participants have working environments and data on their machines. (20 min) Introduction to Natural Language Processing and Sentiment Analysis (15 min) Importing and preprocessing the textual data (15 min) Exercise 1: Toy example to demonstrate how sentiment analysis works (30 min) Exercise 2: Implementing a sentiment analysis model using real data