1/1/2018 - How to Perform a Cognitive Keyword Extraction

How to Perform a Cognitive Keyword Extraction On Unstructured Text


In this blog post, I will demonstrate how to perform a cognitive keyword extraction using Flow.

Flow provides direct interfaces to state-of-the-art cognitive computing platforms. These interfaces combined with Flow's generic data and flexible workflow development environment allow you to invoke advanced artificial intelligence functions with no code.

This post will provide a worked example showing how to perform a keyword extraction on unstructured text data by using the IBM Watson cognitive computing actions.

The example we will explore will show how to take in unstructured text strings from a target dataset and return a new dataset which contains the results of Watson's cognitive analysis. The resulting dataset will hold all extracted keywords, the text sentiment towards those keywords, and an emotion distribution for each keyword.

You will need to set up IBM Watson Cognitive credentials to perform this example. You can learn how to set up these credentials in the blog post here. Watson provides a "lite" plan which allows up to 30,000 natural language calls per month for free.

If you do not have a Flow account - register here. Flow is free for personal use.

To learn how to perform a cognitive keyword extraction in Flow - check out the video below: