Presentation: Practical NLP for the Real World

Most companies have an abundance of text data they could leverage information from. Whether you'd like to classify incoming support ticket requests, detect when a candidate is available for an interview or suggest useful replies to emails, leveraging NLP techniques can often improve your products and allow you to build entirely new ones.

The field of NLP is often not the most approachable, however. The field ranges from linguistics to cutting edge deep learning, so it can be hard to find tools to build a practical product in a reasonable timeframe.

In this talk, we will cover concrete examples of how to build practical applications using NLP. In the real world, most gains come from improvements to the pipeline, not necessarily the model. For this reason, we will dive into data visualization and labelling, as well as model validation.

We will walk through code, plots, and leave ample room for practical questions.