1Z0-184-25 Question 154
Single answerYour organization manages a large volume of scanned invoices stored in OCI Object Storage. You need to automate extracting the text and key invoice details (e.g., vendor name, invoice number, total amount) for downstream analysis in an Oracle Autonomous Database. Which approach should you implement to leverage Oracle Cloud Infrastructure� AI services for text extraction and entity recognition?
- A
Use OCI Vision to perform OCR on the images, then feed the extracted text into OCI Language for entity extraction.
- B
Deploy a custom TensorFlow model on OCI Data Science for image-based classification and manually parse the text within the model code.
- C
Use OCI Vision� custom classification capability to directly label invoice fields and omit the need for NLP parsing.
- D
Write a custom script that directly reads images from OCI Object Storage and uses standard libraries to parse text, bypassing Oracle AI services.
Show answer and explanation
Correct answer: A
Explanation
The recommended solution for invoice processing uses OCI Vision� OCR functionality followed by entity extraction via OCI Language. This architecture aligns with Oracle� best practices for leveraging multiple AI services: Vision for text extraction from images, and Language for NER (Named Entity Recognition) or field extraction. For more information, refer to the OCI Vision and OCI Language documentation in Oracle Cloud Infrastructure� official documentation.
- A. Correct.
Option 1 is correct. Oracle Cloud Infrastructure Vision can detect and extract text (OCR) from images, which you can then send to OCI Language for entity extraction. This two-step approach is a best practice for handling scanned documents, because Vision specializes in OCR, while Language specializes in natural language processing. The structured data can then be stored and analyzed in your database.
- B. Incorrect.
Option 2 is incorrect. While you could build a custom model with OCI Data Science, you would be reinventing the OCR process rather than leveraging OCI Vision, which is specifically designed for text extraction from images. Using specialized AI services typically yields higher accuracy and reduces development effort.
- C. Incorrect.
Option 3 is incorrect. OCI Vision� custom classification capability is generally used for identifying classes of images rather than extracting text from documents. You would still need a natural language processing service to parse invoice fields from the text.
- D. Incorrect.
Option 4 is incorrect. Bypassing Oracle AI services and attempting to build or manage OCR and entity extraction logic yourself is more error-prone. It also underutilizes the pre-built AI services Oracle Cloud provides, which significantly streamline these tasks.