/amazon-comprehend-custom-entity

Notebooks and recipes for creating custom entity recognizer for Amazon comprehend.

Primary LanguageJupyter NotebookMIT LicenseMIT

Notebooks and recipes for creating custom entity recognizer for Amazon comprehend.

Here are below the companion notebooks for the blog post called "Build a custom entity recognizer using Amazon Comprehend" https://aws.amazon.com/blogs/machine-learning/build-a-custom-entity-recognizer-using-amazon-comprehend/

Notebooks:

  • 1-AWS-Comprehend-Custom-Entities.ipynb
  • 2-BlazingText-Word2Vec-Telco-tweents.ipynb
  • 3-AWS-Comprehend-Negative-Custom-Classifier.ipynb

First Notebook shows you how to train a custom entity recognizer from a list of smartphone devices.

Second Notebook shows you how to train a Word2Vec model using the twitter dataset.

Third Notebook leverages the keywords generated from the second notebook to create a custom sentiment classifier.