Create a Custom Skill For Azure AI Search (Part 2)

 

Custom Text Classification Skill

You can map a text passage to various user-defined classes using custom text classification. To automatically determine a book's genre, for instance, you could train a model on the synopsis included on the back cover. You then add a genre facet to your online store search engine using that defined genre.

You need to consider the following to enrich a search index using a custom text classification model:

• Store your documents so they can be accessed by Language Studio and Azure AI Search indexers.

• Create a custom text classification project.

• Train and test your model.

• Create a search index based on your stored documents.

• Create a function app that uses your deployed trained model.

• Update your search solution, your index, indexer, and custom skillset.

Store Your Data

Language Studio and Azure AI Services both provide access to Azure Blob storage. The easiest option is to select Container as the container must be available, although private containers can also be used with some extra setup.

You also need a method for classifying each document in addition to your data. You can manually categorize each document one at a time using the graphical interface that Language Studio offers.

You can choose between two different types of project-

• If a document maps to a single class use a single label classification project.

• If you want to map a document to more than one class, use the multi label classification project.

If you don't want to manually classify each document, you can label all your documents before you create your Azure AI Language project. This process involves creating a labels JSON document.

Conclusion

We have successfully learnt about custom text classification skill.