Azure Document Intelligence Model for Expense Claims
You are building an app that will process scanned expense claims and extract and label the following data: • Merchant information • Time of transaction • Date of transaction • Taxes paid • Total cost You need to recommend an Azure AI Document Intelligence model for the app. The solution must minimize development effort. What should you use?
Community Votes
100% of anonymous learners picked answer D. Votes are pick records left by other test-takers — they are not the verified answer.
Community Insight
This question tests the ability to map required document extraction fields to a prebuilt Azure AI Document Intelligence model, avoiding the trap of choosing a custom model when a zero-training prebuilt option exists.
Selecting the right Azure AI Document Intelligence model for extracting data from scanned expense claims requires matching the required fields to a prebuilt model to minimize development effort. This page establishes that the prebuilt receipt model natively extracts the exact fields requested, making it the optimal choice.
Choosing a custom template or neural model (B or C) under the assumption that specific field requirements necessitate custom training, failing to recognize that the prebuilt receipt model already supports these exact fields natively.
Community Discussion (5 comments)
Comments & Corrections
No comments yet — spotted an error or have a note? Share it below.
Expert Analysis
Why the Answer Is Correct
The prebuilt receipt model in Azure AI Document Intelligence is specifically designed to extract key entities from sales receipts and expense claims. It natively supports extracting merchant information, transaction date and time, taxes, and total cost without any training data. By using this prebuilt model, developers achieve the required extraction functionality with zero custom labeling, perfectly satisfying the requirement to minimize development effort.Why the Other Options Are Wrong
The prebuilt Read model (Option A) only extracts printed and handwritten text as raw lines and words without any semantic understanding or key-value pairing, so it cannot identify specific fields like "Taxes paid". Custom template (Option B) and custom neural (Option C) models require you to provide training data, label the fields manually, and train the model, which significantly increases development effort compared to an out-of-the-box prebuilt model.Community Comment Notes
The community unanimously agrees that the prebuilt receipt model is the correct choice for this scenario. As user syupwsh pointed out by providing the official concept documentation, the prebuilt receipt model is explicitly designed for this exact use case. User JakeCallham noted it is "Obviously D" due to the direct match between the required fields and the prebuilt capabilities.Official Reference
Exam Strategy
When asked to extract specific key-value pairs from standard business documents while minimizing effort, always look for a matching prebuilt model first. Custom models should only be chosen if no prebuilt model covers the specific fields required by the application.