ExamTopics Logo
Mail Us[email protected]
Menu
Microsoft Discussions
exam questions

Exam AI-102 All Questions

View all questions & answers for the AI-102 exam
Go to Exam

Exam AI-102 topic 4 question 18 discussion

Actual exam question from Microsoft's AI-102
Question #: 18
Topic #: 4
[All AI-102 Questions]

HOTSPOT -

You have an Azure subscription that contains an Azure AI Document Intelligence resource named DI1.

You build an app named App1 that analyzes PDF files for handwritten content by using DI1.

You need to ensure that App1 will recognize the handwritten content.

How should you complete the code? To answer, select the appropriate options in the answer area.

Show Suggested Answer Hide Answer
Suggested Answer:
by audlindr at Feb. 29, 2024, 8:44 p.m.

Comments

Chosen Answer:
This is a voting comment (?). It is better to Upvote an existing comment if you don't have anything to add.
Switch to a voting comment New
Submit Cancel
Mehe323
Highly Voted 1 year, 3 months ago
The first answer should be read: Read OCR - Extract print and handwritten text including words, locations, and detected languages. https://learn.microsoft.com/en-us/azure/ai-services/document-intelligence/concept-model-overview?view=doc-intel-4.0.0
upvoted 17 times
...
AlviraTony
Highly Voted 1 year, 3 months ago
Prebuilt-read is correct which can classify the test extracted as handwrittern or printed
upvoted 10 times
...
syupwsh
Most Recent 4 months, 3 weeks ago
"prebuilt-read" is CORRECT. This model is designed to extract both printed and handwritten text from documents such as PDFs. Since your app needs to recognize handwritten content, the "prebuilt-read" model is the appropriate choice. It effectively identifies text from various sources, including handwritten notes, making it ideal for your use case. 0.75 is CORRECT. This threshold is commonly used in AI document analysis tasks to strike a balance between precision and recall. A confidence level of 0.75 means the model is 75% confident that the detected content is handwritten, which is generally reliable for this type of task. It minimizes false positives while ensuring accurate detection of handwritten content.
upvoted 4 times
...
famco
9 months, 3 weeks ago
read model supports handwritten (I'm too tired to check if prebuilt-document also supports handwritten. I do not know why not). https://learn.microsoft.com/en-us/azure/ai-services/document-intelligence/concept-read?view=doc-intel-4.0.0&tabs=sample-code#handwritten-style-for-text-lines
upvoted 2 times
famco
9 months, 3 weeks ago
of course, prebuilt-document also supports it as seen in this code example: https://learn.microsoft.com/en-us/dotnet/api/overview/azure/ai.formrecognizer-readme?view=azure-dotnet#use-the-prebuilt-general-document-model Microsoft has no quality or taste.
upvoted 2 times
famco
9 months, 3 weeks ago
but considering this code extract is the one that is taken and there are no samples with prebuilt-read, and the Microsoft guy did not see the read one, I will go for prebuilt-document.
upvoted 1 times
famco
9 months, 3 weeks ago
Nope. That path is also not there. There is an example for the read one as well with handwritten recognized: https://learn.microsoft.com/en-us/dotnet/api/overview/azure/ai.formrecognizer-readme?view=azure-dotnet#use-the-prebuilt-read-model
upvoted 1 times
famco
9 months, 3 weeks ago
And exactly the same piece of code and in the same page. Now how can I guess which one the great employee from Microsoft saw when creating this wonderful question.
upvoted 1 times
famco
9 months, 3 weeks ago
And why not 0.1 as the answer to the next question. I want all more than 10 percent. Who said I cannot do that?? Such substandard questions
upvoted 1 times
...
...
...
...
...
...
anto69
10 months, 3 weeks ago
ChatGPT: prebuilt-read
upvoted 1 times
...
krzkrzkra
11 months, 3 weeks ago
1. "prebuilt-document" 2. 0.75
upvoted 3 times
...
rookiee1111
1 year ago
prebuilt-read that is the correct one https://learn.microsoft.com/en-gb/training/modules/use-prebuilt-form-recognizer-models/3-use-general-document-read-layout-models
upvoted 3 times
...
reiwanotora
1 year, 1 month ago
1. "prebuilt-document" 2. 0.75
upvoted 2 times
...
Murtuza
1 year, 2 months ago
The given answers by exam topics is CORRECT
upvoted 3 times
...
Murtuza
1 year, 3 months ago
It appears that the code snippet you’ve provided is in C#. Let’s break it down You’re iterating through the DocumentStyle objects within the result.Styles You’re invoking an AnalyzeDocumentOperation from an AnalyzeDocumentFromUriAsync method call. The operation is awaited until completion For each style, you’re checking if it’s handwritten (based on the IsHandwritten property) and if its confidence level is greater than 0.1
upvoted 1 times
...
audlindr
1 year, 4 months ago
Reference: https://learn.microsoft.com/en-us/dotnet/api/overview/azure/ai.formrecognizer-readme?view=azure-dotnet#use-the-prebuilt-general-document-model
upvoted 3 times
Mehe323
1 year, 3 months ago
It is read: Read OCR - Extract print and handwritten text including words, locations, and detected languages. https://learn.microsoft.com/en-us/azure/ai-services/document-intelligence/concept-read?view=doc-intel-4.0.0
upvoted 3 times
...
Harry300
1 year, 4 months ago
why document and not the read model? the code appears in both examples on that website
upvoted 7 times
JakeCallham
10 months, 1 week ago
document inherits from read, but if you only need ocr, read is more sufficient.
upvoted 1 times
...
...
...
Community vote distribution
A (35%)
C (25%)
B (20%)
Other
Most Voted
A voting comment increases the vote count for the chosen answer by one.

Upvoting a comment with a selected answer will also increase the vote count towards that answer by one. So if you see a comment that you already agree with, you can upvote it instead of posting a new comment.

SaveCancel