Building a Sentiment Classifier Using Create ML to Classify User Reviews
In the previous chapter, I have walked you through the basics of Create ML and showed you how to train an image classifier. As mentioned, Create ML doesn't limit ML model training to images. You can use it to train your own text classifier model to classify natural language text.
What we are going to do in this chapter is to create a sentiment model for classifying product and movie reviews. The trained model takes in opinions in natural language text, analyze, and classifies it into positive/negative reviews. Furthermore, I will show you how to compile and test the model in Playgrounds.
Before you move on, please make sure you check out the previous two chapters. I assume you already equip yourself with the basics of Create ML and Core ML.
To continue reading and access the full version of the book, please get the full copy here. You will also be able to access the full source code of the project.