AI 900 exam question - Part 2

                                           scan from scanner to read article.

31)  which service should you use to extract text , key/Value pairs and table data automatically from scanned documents?

   a) Form recognizer 

   b) Text Analytics

   c) Language Understanding

   d) Custom vision

32 )You want to use computer vision service to identify the location of individual items in an image. Which of following feature should you retrieve ?

   a)object

   b)Tags

   c)Categories

34 ) You want use the computer vision service to analyse image and to use language service to analyse text.It is a must requirement that developer only use one key and endpoints to access all of your services, What kind recourses should you create in your azure subscription ?

   a)Computer vision

   b)Cognitive services

    c)Custom vision

35 )You use a azure Machine Learning designer pipeline to train and test a binary test a binary classification model.You review the model's performance metrics in an evaluate model module and note that it has AUC score of 0.3. what can you conclude about the model ?

   a)The model can explain 30% of variance between true and predicated labels.

   b) The model predicates accurately for 70% of test cases

  c) The model performs worse than random guessing.

Note :The ROC AUC score can range from 0 to 1.

A score of 0.5 indicates random guessing,

score of 1 indicates perfect performance.

A score slightly above 0.5 shows that a model has at least "some" (albeit small) predictive power. This is generally inadequate for any real applications.

above 0.8 is considered good,

score above 0.9 is considered great. 

36) You use Azure Machine Learning designer to create a training pipeline for classification model. What must you do before deploying the model as a service ? a) Create an interface pipeline from the training pipeline b) Add an Evaluate Model module to training pipeline c) Clone the training pipeline with a different name

37) You are using Azure Machine Learning designer to create training pipeline for binary classification model.you have added dataset contain feature and labels, a two class decision forest module and train module.you plan to use Score Model and Evaluate Model modules to test the trained model with subset of dataset that was not used for training. what is another module should you add ? a) join data b) Split data c) Select column in dataset

38)The ability to extract subtotal and total from receipt is a capability of the A)custom vision b)Form recognizer c)Ink Recognizer d) Test Analytics

39)You are building a tool that will process images from retail store and identify the products of competitors.The solution will use a custome model.which azure cognitive service should you use ? a)custom vision b)form recognizer c)face d)Computer vision

40)When you design an AI system to assess wheather lonas should be approved, ha factor used to make the decision should be explinable. This is an example of which Microsoft guiding principle for Repsponsible AI ? a)transparency b)inclusiveness c)Fairness d)Private an security

AI 900 exam question - Part 1

scan above image read this article.

Like page and share link to others

1) Which AI Service should you use to create a bot from a frequently asked question (FAQ) document ?
a) Qna Maker
b) Language understanding (LUIS)
c) Test Analysis
d) Speech text


2) You build machine learning model by using automated machine learning user interface (UI). you need to ensure thta model meet the microsoft transparency principle for responsible AI what should you do ?
a) Set vaildation type to Auto.
b) Enable Explain best model
c) Set Primary metic to accurency.

3)Which scenario is an example of webchart bot?
a) Determine whether review entered on website for concert are positive or negative, and then add thumbs up or thumps down emoji to reviews. b) Translate into English questions entered by customer at koisk so that appropriate person can call the customer back.
c) Accept questions through email , and then route the email messages to correct person based on content messages.
d) From Website interface , answer common question about scheduled events and ticket purchases for music festival.

4) For machine Learning Process , how should you spilt data for training and evaluation ?
a) Use features for training and Labesl for evalution.
b) Randomly split the data into rows for trainig and rows for evalution.
c) Use labels for training and features for evolution.
d) Randomly split the data into column for training and columns for evaluations.


5)Which of following is calculated probity of correct image classification

A)Accuracy
b)Confidence
c) Root Mean Square Error.
d) Sentiment

6)Ensuring AI system dows not provide a prediction when important fields conation unusual or missing values is *** principle for responsible AI.
a) an inclusiveness
b) a privacy and security
c) a reliability and safety
d) a transparency


7) Ensuring that numeric variable in variable data are on similar scale is example of *** **

a) Data ingestion
b) Feature engineering
c) Feature selection
d) Model training

8) You want to create a model to predict sales of ice cream based on historic data that includes daily ice cream sales total and weather measurement. what Azure service should you use ?
a) Azure Machine Learning
b) Azure Bot.
c) Language

9) The interactive answering of question entered by user as part an application is an example of
a)anomaly detection
b) computer vision
c)conversation AI
d)forecasting

10)You are designing an AI Application that uses images to detect creks in windows-panes of building and warns when windows pane should be repaired or replaced. What AI workload is described ?
a)Computer Vision
b) Natural Language Processing.
c) Anomaly Detection

13 >Match the column
Regression          Predict How many minutes a late flight will arrive based on the amount of snowfall at an airport
clustering        Segment customer into different group to support a marketing department
Classification      Predict wheatear a student will complete a university course

14) A Predictive app provide audio output for visually impaired used.which principle of Responsible AI is reflected here ?
a) Transparency
b) inclusiveness
c) Fainness

15) You need to determine the location of cars in an image so that can estimate the distance between that cars. Which type computer vision should you use ?
a) optical character recognition (OCR)
b)object detection
c) Image classification
d) face detection.

 
16) Azure Machine Learing desinger lets you create machine learning models by:

  a) Adding and connecting modules on visaul canvas. 

b)automatically performing common data preparation tasks.
c)automatically selecting an algorithm to build the most accurate model.
d) using a code-fiart notebook experience.


17) Your company is exploring the use of voice recognition technologies in its smart home devices.The company wants to identify any barriers that might unintentionally leave out specific user groups. This is example of which Microsoft guiding principle for responsible AI?
a) Accountability
b) fairness
c) inclusiveness 

d) Private and security.

18) A housing dealership wants to use historic apartment sales data to train a machine learning model. The model should predict the price of pre-owned apartment based on its location, size and apartment aging. what kind of machine model should dealership use automated machine learning to create ? 

a) classification
b) regression
c) Time series forecasting

20) What are three Microsoft guiding principles for responsible AI >

 a) Knowledgeability

b)decisiveness
c) inclusiveness
d) fairness
e)reliability and safety

21> A bank wants to use historic credit card repayment records to categorize credit card application as low-risk or high risk based on characterises like the credit amount, the income of borrower and credit peroid. What kind of machine learning model should bank use automated machine learning to create ?
a)Classification
b)Regression
c)Time series forecasting

Note ::           Numeric ---> Regression
                      Category ---> Classification

23) Predicating how many hours of overtime a delivery a delivery person will work based on number of order received is example of
a) classification
b) Clustering
c)Regression

24) Which following service should you use to train an object detection model by using own images.

a) Computer vision

b) Custom vision 

c)Form Recognizer

d)Video indexer

25) You are creating training pipeline for regression model. You use a dataset has multiple numeric column in which values on different scales. you want to transform the numeric column so that values are all on similar scale.You also want reformation to scale relative to maximum and maximum values in each column. Which module you add to pipeline?

a) Select Column in Dataset

b)Normalise Data.

c)Clean missing Data.

26)What are two tasks that can performed by using Computer Vision service ?

a) Trian A custome image classification model.

b) Detect faces in an image.

C) Recognise hadwritten text.

d) Translate the text in an image between languages.

27)You are working in a tour & tourism company.Your company wants to use the Computer Vision service to analyse image of location and identify well-known monuments. What should you do ?

a) Retrieve the objects in the image

b)Retrieve the categories for image, specifying the celebrations domain

c)Retrieve the categories for image, specifying the landmarks domain

28) Why do you spilt data into training and validation sets ?

a) Data is split two sets in order to create two models, one models with trining set and a different model with validation set.

b) Splitting data into two set enables you to compare the labesl that model redicits with the actaul known lables in original dataset.

c)Only split data when you use Azure Machine Learing Designer ,not in other machine learing scenarios. 

29)What are two metrics that you can use to evaluate a regression model ?

a)coefficient of determination (R2)

b) root mean squared error(RMSE)

c) F1 score

d) area under cureve (AUC)

e) balanced accuracy

30)You are processing photos of runners in race.You need to read the numbers on runners shirts to identify the the runners in photoes.Which type of computer vision should you use ?

a) facial recognition

b) optical character recognition (OCR)

c) semantic segemention

d) object detecton