AI-102 exam dumps: questions 401 to 492 of 492

Correct answers: A, C, E

• A. Correct.

  The prebuilt Invoice model is specifically designed to extract common fields from invoices, such as invoice numbers, dates, and totals. This is the correct model to use for this task.

• B. Incorrect.

  Manually labeling and training a custom model is not necessary when using prebuilt models, as they are already trained for specific document types like invoices.

• C. Correct.

  Azure Form Recognizer can process documents stored in Azure Blob Storage. To use the service efficiently with a large set of documents, connecting to Azure Blob Storage is a best practice.

• D. Incorrect.

  The prebuilt Receipt model is designed for extracting information from receipts, not invoices. Using this model would not provide the desired results for invoice data extraction.

• E. Correct.

  The results from the prebuilt Invoice model must be analyzed to extract the specific fields of interest, such as invoice numbers, customer names, and total amounts.

Correct answer: A

• A. Correct.

  Azure Form Recognizer with its prebuilt Invoice model is designed specifically to extract key-value pairs, tables, and other structured data from invoices and other documents. It minimizes the need for custom development or training.

• B. Incorrect.

  Azure Custom Vision is used for image classification and object detection, not for extracting structured data from documents.

• C. Incorrect.

  Azure Cognitive Search is primarily designed for indexing and searching data, not for extracting structured data from documents like invoices.

• D. Incorrect.

  Azure Machine Learning Studio is a platform for building, training, and deploying custom machine learning models, but it requires significantly more time and effort compared to using prebuilt models like Azure Form Recognizer.

Correct answer: A

• A. Correct.

  The Form Recognizer prebuilt Invoice model is specifically designed to extract key fields from invoices, such as invoice number, date, total amount, and vendor name, without requiring custom training.

• B. Incorrect.

  Azure Cognitive Search with skillsets can process documents and extract text, but it is not tailored for extracting structured fields like invoice-specific data.

• C. Incorrect.

  A custom Form Recognizer model would require you to label and train the model with your own data, which is unnecessary in this scenario since the prebuilt Invoice model already meets the requirements.

• D. Incorrect.

  Text Analytics for entity recognition can extract named entities from text, but it is not optimized for structured data extraction from invoices.

Correct answer: C

• A. Incorrect.

  Testing the model is important but it can only be done after the model has been trained. This step is not performed immediately after labeling the data.

• B. Incorrect.

  Publishing the model is a required step for production use, but it must be trained first before it can be published.

• C. Correct.

  Training the model using labeled data is the next step after labeling in Form Recognizer Studio. This process enables the model to learn how to extract the specified fields.

• D. Incorrect.

  Configuring a Cognitive Services resource is necessary for using Form Recognizer, but this step is typically done before labeling and training the model.

Correct answers: A, B, D, E

• A. Correct.

  Correct. Labeling the sample documents with specific fields and regions is essential for training a custom document intelligence model in Azure Form Recognizer.

• B. Correct.

  Correct. The training data must be uploaded to Azure Blob Storage and connected to the Form Recognizer service for the model to access it during the training process.

• C. Incorrect.

  Incorrect. Azure Cognitive Search is not required for preprocessing documents when creating a custom document intelligence model in Form Recognizer.

• D. Correct.

  Correct. Training the model is a critical step, and it can be done using the Form Recognizer Studio or through the REST API.

• E. Correct.

  Correct. Evaluating the model's accuracy with test documents and refining the training data ensures the model performs well in production scenarios.