Text Classifiers

Here is a short introduction to text classifiers.

What is a text classifier?

A text classifier answers the question To which predefined class (category) is this text most likely to belong? For example a classifier trained on web categories can answer "I am by 99% certain that the web page belongs to the category food."

Can I create my own classifiers?

Yes! Just sign up and create a classifier, add classes and train each class on documents. You can upload files for each class.

What does it mean to train a classifier?

Before it can classify anything will require some training. Training involves creating all wanted classes (categories) and training each class on text documents that are of that type. For example creating a web page categorizer would require that some classes are created (food, cars, news, health etc). Then each class is trained on documents that are of that type. E.g. putting texts about food into the food category, car texts into the car category and so on.

Can I use the public classifiers in my project?

Yes! There are plenty of classifiers that already has been created, such as Sentiment and Topics. You can access them via our APIs.