How to build high-quality Classifiers with limited training data

Posted by

Automate Classifiers are the bread and butter of automated content understanding. They help categorize documents against any desired taxonomy, from IPTC Newscodes to sentiment, intent, stance, or emotion. Building your own classifier with Factory no-code AI is as easy as it gets (and secure, too!), but what happens if you don’t have enough machine learning training data?

This presentation describes how using Factory Transformer-based classifier trainer can help you get started with limited training data, and then build a more robust-high quality engine with the RNN (Recurrent Neural Network) classifier trainer.