Exam PL-200 topic 5 question 17 discussion

The first one should ONLY be Text recognition, because it is reading the text from a PDF and recognizing the text.... Key phrase extraction would not make sense because it is NOT a goal to identify key-features in the text... The second one CANNOT be Prediction model, because this is NOT a prebuilt-AI model. The question clearly asks for a PRE-BUILD AI model. Therefore it can only be Sentiment analysis

for me.. its 1. Text recognition model 2. Prediction Model https://docs.microsoft.com/en-us/ai-builder/prebuilt-text-recognition https://docs.microsoft.com/en-us/ai-builder/prediction-overview

Copilot To integrate prebuilt AI Builder models with Power Automate flows for the scenarios you described, you should use the following models: Extract key phrases from a PDF document: Key Phrase Extraction: This prebuilt model identifies important phrases in text, making it suitable for extracting key phrases from PDF documents. Determine the likelihood that customers will purchase additional products: Prediction Model: This model can be used to predict outcomes based on historical data. For determining the likelihood of customers purchasing additional products, you can use a prebuilt prediction model. These models can be easily integrated into Power Automate flows to automate and enhance your business processes.

Extract key phrases from a PDF document: C. Text recognition model and key phrases extraction model. First, use the Text recognition model to extract text from the PDF document, and then apply the Key phrase extraction model to identify key phrases from the extracted text1. Determine the likelihood that customers will purchase additional products: D. Prediction model. The Prediction model in AI Builder is designed to analyze patterns in historical data and predict future outcomes, such as the likelihood of customers purchasing additional products

How on earth can a sentiment analysis model predict the likelihood of a customer to buy additional products? I mean, e.g. to select all customers where the likelihood is over 93.2% ? The only reason is a potential hypothesis that happy customers may buy again. But it is not what the sentiment analysis does (eg in a Monopoly an very unhappy customer is highly likely for re purchase. In tourist places with high walk-in customers the extremely happy customer is unlikely to buy again as he might have gone home). But i also see the point that prediction models are not prebuilt. Thanks to all the comments, i know now what the expected answer should be.

answer is correct.

3- text recognition model and key phrase extraction model 4- prediction model

1: Text recognition model and key phrase 2: Sentiment analysis https://learn.microsoft.com/en-us/ai-builder/prebuilt-overview

Text recognition + key phrase extraction. Exactly showed in the doc as you need to run text recognition on PDF file first. Sentimental analysis - I chose prediction initially but realised it was not a pre-build model. Sentimental analysis is, and does the job

Q1: key phrase extraction --> text recognition uses OCR not using a text file Q2: sentiment analysis is pre-built and can be used on the pdf

You can find in this link the list of pre-built AI Models: https://learn.microsoft.com/en-us/ai-builder/prebuilt-overview So I think the answer is: 1. Key phrase extraction 2. Category classification (cannot be sentimental analysis because it is appliable on a text)

1-in this link it says that key phrase is a prebuilt model https://docs.microsoft.com/en-us/ai-builder/prebuilt-key-phrase 2-https://learn.microsoft.com/en-us/ai-builder/prebuilt-text-recognition it says that"For PDF documents, only the first 2,000 pages are processed." which makes the answer correct I think.

Question 1 is Text Rec AND Key Phrase Extraction. The question clearly says "EXTRACT KEY PHRASES". Because it's a PDF, you must first use Text Recognition, and then ALSO use Key Phrase Extraction. Text Recognition alone WILL NOT extract key phrases, instead, it's just a data prep step to enable the Key Phrase Extraction model to do its job. Second should be Sentiment Analysis. Prediction would be better, but is not a pre-built model, so doesn't satisfy the constraints given in the question.

I think the first one should be Key phrase extraction model as it is designed to extract key phrases as specified. The text recognition model only reads hand-written notes or printed content when you take a pic of it. For the second one, it should be Sentiment Analysis.

For me is: 1. Just Key Phrase Extraction model and 2. Sentiment analysis model. PS. For Q1 >> The question specifically asks only for key phrases, not some other text.

I think right answers are : -Text recognition model -Sentiment analysis model Check this link for prebuilt AI model list (that is requested in the question) : https://docs.microsoft.com/en-us/ai-builder/prebuilt-overview Prediction are not prebuilt models but custom...

Text recognition AND key phrase Sentiment analysis