Technology Blogs by SAP
Learn how to extend and personalize SAP applications. Follow the SAP technology blog for insights into SAP BTP, ABAP, SAP Analytics Cloud, SAP HANA, and more.
Showing results for 
Search instead for 
Did you mean: 

Document Information Extraction

With the smart capabilities of Document Information Extraction (which is part of the SAP AI Business Services Portfolio), information can be easily extracted from business documents using the pre-trained AI models.

The pre-trained models can extract all the relevant information from standard documents (such as invoices, purchase orders, payment advices and business cards). However, in our day to day use, we come across various types of business documents which might also have special formats and custom fields (for example power of attorney). Such documents are sometimes unrecognizable for the pre-trained models.

In this blog we would like to share how you can add a customization touch to your information extraction process by using templates based on your specific business needs.

Template Feature

We would like to highlight the template feature here, which is one of the key features of Document Information Extraction UI. It supports users to extract information very easily and flexibly from many different types of documents. With the template feature, users are able to create, reuse, edit, and delete templates, based on schemas and sample document files.

Note: Schemas are a basis for creating templates. Users can select schemas and associated templates when adding documents.

As we already mentioned above, Document Information Extraction uses pre-trained models to extract information from your documents. Templates can be used to extend these existing pre-trained models to yield even greater benefit for you. This feature offers two scenarios for you:

Fine-tuning, where you can use existing schemas (which are used for pre-trained models), for creating new templates and use them to extract information from standard document types more precisely and accurately

Full customization, where you can create new schemas and new templates to extract information from custom document types

We have prepared a short demo video to give you a first idea about how to create your own template using an existing schema. Let us know in the comments below if you would be interested to see more such demo videos, also for the schema creation.

Document Information Extraction offers the template feature in many different languages so that you can enjoy the customization of documents fully and more flexibly. Find here the list of languages available.

Now you can use the template feature of Document Information Extraction to customize and enhance the information extraction process. Find more information on the SAP Help Portal page for Document Information Extraction.

Here are our recommendations for using templates: Using Templates: General Recommendations.

What’s more?

Document Information Extraction also offers the template feature in free tier. Read more about it here. Stay tuned for more updates on new features.

We recommend to also read our recent blogs about of Document Information Extraction:

For more information about the Document Information Extraction please refer to the pages linked below.

Learn more

Read more about the news of Document Information Extraction on the help portal!

What is Document Information Extraction?

Document Information Extraction is one of the SAP AI Business Services on the SAP Business Technology Platform (SAP BTP). This ML-enabled service is available through the Cloud Platform Enterprise Agreement (CPEA) and also in the Pay-As-You-Go (PAYGO) model.

SAP Community Pages:

Tutorials & Learnings

Blog posts: