Extract Information with Azure Content Understanding in Foundry Tools AI-102 Practice Questions (Page 1)

Questions 1–5

1. 1application · medium

   A financial analyst receives monthly statements in PDF format that contain a table of transactions with columns for date, description, and amount. They need to extract the table data into a structured format for further analysis, and also extract the account number from the header. What should they configure in Azure Content Understanding?

   • AA table extraction pipeline with OCR enabled, and add a custom field for the account number.
   • BA summarization pipeline that summarizes the transactions and includes the account number in the summary.
   • CAn entity extraction pipeline that extracts all dates and amounts, and then manually reconstructs the table.
   • DA document classification pipeline that categorizes the statement as a monthly report and then uses the category to identify the table.
   Reveal Solution Discussion Select an answer first
2. 2application · medium

   A financial audit firm receives monthly statements that contain a large table of transactions, a summary paragraph, and a footer with the account number. They need to extract the transaction table, the account number, and the summary text. Using Azure Content Understanding in Foundry Tools, what should they configure?

   • AA document classification pipeline that categorizes the statement and extracts the table.
   • BAn image extraction pipeline that saves the table as an image for manual review.
   • CA custom extraction pipeline with table extraction, an attribute for account number, and OCR for the summary text.
   • DA summarization pipeline that summarizes the table and includes the account number.
   Reveal Solution Discussion Select an answer first
4. 3application · medium

   A research team has a collection of scientific papers in PDF format that contain figures, charts, and diagrams. They need to extract all images from the papers and also capture the figure captions as text. What should they configure in Azure Content Understanding?

   • AA document classification pipeline that categorizes the papers and then uses the category to find images.
   • BA table extraction pipeline that extracts any tables in the papers and assumes the figures are included.
   • CAn image extraction pipeline that extracts images and also uses OCR to capture the text in the captions.
   • DA summarization pipeline that summarizes the paper and includes the figure captions in the summary.
   Reveal Solution Discussion Select an answer first
5. 4expert · hard

   A utility company processes meter reading forms that are filled out by hand. The forms include a table of meter readings and a section for the technician's ID and the date. The handwriting is often messy, and the table layout can vary slightly. They need to extract the technician ID, the date, and the meter readings with high accuracy. What should they do in Azure Content Understanding?

   • ACreate a custom OCR pipeline with custom field extraction for technician ID and date, and a custom table extraction for meter readings, then train with a diverse set of handwritten forms.
   • BUse a document classification pipeline to categorize the form and then use the category to infer the technician ID and meter readings.
   • CUse an entity extraction pipeline to extract all numbers and then manually match them to the technician ID and meter readings.
   • DUse a pre-built OCR model and assume it will handle all handwriting and table layouts without customization.
   Reveal Solution Discussion Select an answer first
6. 5foundation · easy

   In Azure Content Understanding, which capability would you use to generate a brief overview of a long research paper?

   • AAttribute detection
   • BTable extraction
   • COCR pipeline
   • DDocument summarization
   Reveal Solution Discussion Select an answer first