Cohere tutorial: Entity extraction

Extracting information from text is a common task in language processing. LLMs can extract entities that are difficult to extract using other NLP methods (and where pre-training provides the model with some context on these entities). This is an overview of using generative LLMs to extract entities

Let's get started

First of all, register to Cohere https://dashboard.cohere.ai/register

After registration you need to head over to the Playground https://os.cohere.ai/playground

Next, we can check out the Cohere Playground.The Cohere Classify Playground is a great tool for testing your ideas and getting started with a project. It has a clean UI and can export your code in multiple languages.

We will choose generate endpoint with default language model.

For this example I will use the Extract entities from Invoices example. And rewrite it just to show how great this model is.

We will extract names from sentences. You can paste the following sentences in the Playground and see how the model extracts the names.

Extract names from sentences.
 
Sentence: Bob went to the market to buy something for lunch
Name: Bob
--
Sentence: Peter Parker was Spiderman in real life
Name: Peter Parker
--
Sentence: Once I have seen Buffy the vampire slayer
Name:

In my test it was correctly generating the answer even with two examples.

And from the playground we can export the code in multiple languages.

Conclusion

Cohere is giving a great solution to extract entities from text with endless possibilities.

Join our AI Hackathons to test your knowledge and build with assistance of our mentors AI based tools to change the world! Check the upcoming ones here: Cohere Thanksgiving Hackathon & Cohere Holiday Hackathon

Thank you! If you enjoyed this tutorial you can find more and continue reading on our tutorial page