A.  

For a given concept what % of cases will the model incorrectly predict.

B.  

For a given concept, what % of cases will the model not detect.

C.  

For a given concept what % of cases will the model correctly predict.

D.  

For a given concept, what % of cases will the model detect.

A.  

Train models using unsupervised learning.

B.  

View, save, and modify dataset model versions.

C.  

Access detailed, queryable charts, statistics, and customizable dashboards.

D.  

Monitor model performance and receive recommendations.

A.  

Data labeling session name and dataset.

B.  

Data labeling name, dataset. and Al Center project.

C.  

Dataset and Al Center project.

D.  

Data labeling name, language, and number of documents.

A.  

Kill.

B.  

Remove and restart.

C.  

Kill, remove, and restart.

D.  

Remove and kill.

A.  

Less than 5 minutes.

B.  

Between 5 and 10 minutes.

C.  

Between 10 and 15 minutes.

D.  

More than 15 minutes.

A.  

The Flow Switch activity is designed solely for usage in sequence workflows.

B.  

Default cases can be numbered.

C.  

Two default cases can be assigned in the Default section.

D.  

The default TypeArgument property for the Flow Switch activity is set to Int32.

A.  

A Pipeline can be scheduled at multiple future dates or with a recurring schedule.

B.  

A Pipeline can be scheduled at a single future date.

C.  

A Pipeline can be scheduled with a recurring schedule.

D.  

A Pipeline can be scheduled at a single future date or with a recurring schedule.

A.  

Use the "Rebalance" training mode to pin the label to more communications.

B.  

Use the 'Teach" training mode to pin the label to more communications.

C.  

Use the "Low confidence" training mode to pin the label to more communications.

D.  

Use the "Search" feature to pin the label to more communications.

A.  

Allows human intervention for validating the classification results or the extraction results from different documents.

B.  

Enables the creation of ML packages and allows their consumption within RPA workflows.

C.  

Allows the deployment, management and improvement of ML models and their consumption within RPA workflows in Studio.

D.  

Permits the creation of ML models, their improvement and their consumption within RPA workflows in Studio.

A.  

It provides a simplified way to manually create Datasets, Pipelines, and Datasets for document extraction.

B.  

It automates the creation of document extractors by automatically creating Datasets, Pipelines, and ML Skills but requires manual creation of Pipelines.

C.  

It allows training document extractors without manually creating Datasets, Pipelines, and ML Skills in AI Center.

D.  

It allows training document extractors without manually creating Datasets and Pipelines, but you need to manually deploy the ML Skills in AI Center.

A.  

It allows users to bypass the need for manually creating Datasets, Pipelines, and ML Skills in AI Center and enables training document classifiers directly within Document Understanding.

B.  

It is a pre-trained Machine Learning model that helps you classify documents by providing a prompt.

C.  

It enables you to directly edit and alter the underlying code of the ML models used for classification, giving the users unprecedented control over the machine learning process in Document Understanding.

D.  

It enables users to manually create Datasets, Pipelines, and ML Skills in Document Understanding itself.

A.  

5

B.  

10

C.  

15

D.  

20

A.  

An ML skill is by default public and can be made private.

B.  

An ML skill is by default private and can be made public.

C.  

An ML skill is by default public and can't be made private.

D.  

An ML skill is by default private and can't be made public.

A.  

Average label performance.

B.  

Coverage.

C.  

Balance.

D.  

Underperforming labels.

A.  

Intelligent Keyword Classifier Trainer, Language Based Classifier Trainer, and Image Based Classifier Trainer.

B.  

Image Based Classifier Trainer, Format Based Classifier Trainer, and Machine Learning Classifier Trainer.

C.  

Machine Learning Classifier Trainer, Language Based Classifier, and Keyword Based Classifier Trainer.

D.  

Keyword Based Classifier Trainer, Intelligent Keyword Classifier Trainer, and Machine Learning Classifier Trainer.

A.  

The process requires exception handling

B.  

High process run time

C.  

The process is prone to human error

D.  

Frequent business logic change

A.  

The flowchart undergoes successful validation with no errors displayed, but an exception will occur during the workflow's runtime.

B.  

The flowchart displays an error indicating that some activities are not connected to the others within the flowchart.

C.  

The flowchart will display an error indicating the presence of unconnected activities, but only when validated using the Workflow Analyzer.

D.  

A warning message is displayed in the Output section of Studio, indicating the presence of unconnected activities.

A.  

Present Validation Station, Wait for Document Validation Action and Resume.

B.  

Orchestration Process Activities.

C.  

Document Understanding Process Activities.

D.  

Create Document Validation Action, Wait for Document Validation Action and Resume.

A.  

Identifies what type of document the robot is currently processing.

B.  

Presents a document processing-specific user interface for validating and correcting automatic classification outputs.

C.  

Empowers the closing of the feedback loop to any classification algorithm capable of learning.

D.  

Retrieves the text from any PDF or image, using, only if necessary, the OCR engine.

A.  

A mandatory step in the extractor configuration that allows choosing which extractors are applied to each field.

B.  

A mandatory step in the extractor configuration that allows choosing which extractors are applied to each document type and field.

C.  

A mandatory step in the extractor configuration that allows choosing which extractors are applied to each document type.

D.  

An optional step in the extractor configuration which allows choosing which extractors are applied to each document type.

A.  

Refine

B.  

Explore

C.  

Discover

D.  

Setup

A.  

Specify an output format to standardize the response.

B.  

Combine arithmetic operations with the question to create a better context.

C.  

Combine complex if-then-else type logic questions to create a better context.

D.  

Request confidence levels for predictions.

A.  

ArrayOf or System.DataType[] as it stores indexed values

B.  

System.Collections.Generic.Dictionary as it stores key-value pairs

C.  

System.Collections.Generic.Dictionary as it stores indexed values

D.  

System.Collections.Generic.List as it stores ordered values

A.  

AI Center is the infrastructure on top of which UiPath Document Understanding digitization runs.

B.  

Document Understanding is the infrastructure on which AI Center digitization runs.

C.  

AI Center is the infrastructure on top of which UiPath Document Understanding machine learning models run.

D.  

Document Understanding is the infrastructure on which AI Center machine learning models run.

A.  

Allows the model to learn from human-validated data.

B.  

Allows the extractor to improve its prediction over time by using better OCR (Optical Character Recognition) engines.

C.  

Allows a human to validate and correct the extracted data.

D.  

Improves the extractor accuracy by learning from the classification result.

A.  

Status not valid.

B.  

Status is Unknown.

C.  

Status is Pending.

D.  

Status is New.

A.  

The Document Understanding project must be linked to Action Center.

B.  

The Document Understanding project must have at least five Document Types defined.

C.  

The Document Understanding project must be linked to AI Fabric.

D.  

The Document Understanding project must be linked to AI Center.

A.  

A collection of raw unlabeled communications data of a similar type, that can be associated with up to 10 datasets.

B.  

The model that we create when training the platform to understand the data in those sources.

C.  

A permissioned storage area within the platform which contains communications and labels.

D.  

A user-permissioned project containing a taxonomy with labels and entities.

A.  

The model makes correct predictions across the majority of the dataset.

B.  

All of the labels in the dataset have at least 25 training examples.

C.  

Many of the labels in the taxonomy have amber or red performance warnings.

D.  

2% of the dataset has been reviewed by a model trainer.

A.  

Each page of the images used for schema configuration.

B.  

Each page of the exported documents saved as an image.

C.  

Each page of the images of logs.

D.  

Each page of the images of exported files.

A.  

Label predictions are made based on the text of the message as well as some metadata properties, while general fields are learnt from the assigned span of text and the context of the text surrounding that span.

B.  

Label predictions rely solely on metadata properties, general fields are learnt from the presence of certain key phrases in the text.

C.  

Labels predictions are based on assigned span of text and entities are predicted solely from metadata properties.

D.  

Labels and general fields are both predicted based on the text of the message, with no consideration given to metadata properties or contextual surrounding text.

A.  

Using Read Taxonomy activity.

B.  

Using DeserializeTaxonomy activity.

C.  

Using Load Taxonomy activity.

D.  

Using DocumentTaxonomy.Deserialize method inside an Assign activity.

A.  

Improves the extractor accuracy by learning from the classification result.

B.  

Allows the model to learn from human-validated data.

C.  

Allows the extractor to improve its prediction over time by using better OCR (Optical Character Recognition) engines.

D.  

Allows a human to validate and correct the extracted data.

A.  

DocumentClassifier

B.  

Document

C.  

DocumentUnderstanding

D.  

DocumentMLPackage

A.  

For structured or semi-structured documents in which layouts of different document providers vary greatly.

B.  

For fixed-layout documents for data extraction, including handwriting recognition and signature detection.

C.  

For simple use cases in which data is always found in a strict, predictable format and context.

D.  

When documents have little to no variation in the document layouts.

A.  

Immediate Panel and Watch Panel

B.  

Locals Panel and Watch Panel

C.  

Locals Panel and Immediate Panel

D.  

Watch Panel and Breakpoint Panel

A.  

The main reason is that all taxonomy fields should have their names translated into all the languages presented in the documents to be processed.

B.  

Different Document Understanding components have support for different languages, so they have to be selected according to the requirements.

C.  

This is a standard check but has no actual effect on the Document Understanding process.

D.  

If an ML (Machine Learning) extractor is used, the model schema fields should be renamed according to the most frequently used language from the documents to be processed

A.  

Invoices, receipts, purchase orders, and medical bills.

B.  

Banking forms, tax forms, surveys, and identity cards.

C.  

Contracts, emails, banking forms, and tax forms.

D.  

Contracts, agreements, and emails.

A.  

OCR (Optical Character Recognition) is a method that reads text from images, recognizing each character and its position, while IntelligentOCR is an enhanced version of it that can also work with noisier input data.

B.  

IntelligentOCR is simply a rebranding of the OCR (Optical Character Recognition), both of them being methods that read text from images, recognizing each character and its position.

C.  

OCR (Optical Character Recognition) is a UiPath Studio activity package that contains IntelligentOCR as an activity used to read text from images, recognizing each character and its position. OCR is widely used in Document Understanding processes.

D.  

IntelligentOCR is a UiPath Studio activity package that contains all the activities needed to enable information extraction, while OCR (Optical Character Recognition) is a method that reads text from images, recognizing each character and its position.

A.  

Model Ratings of "Good" or better and individual performance factors rated as "Good" or better.

B.  

Model Ratings of "Good" and individual performance factors rated as "Excellent".

C.  

Model Ratings of "Excellent" and individual performance factors rated as "Good" or better.

D.  

Model Ratings of "Excellent" and individual performance factors rated as "Excellent".

A.  

The number of occurrences and the extraction confidence.

B.  

The page number from which the field was extracted and the exact position on the page.

C.  

The extraction confidence and the digitization confidence.

D.  

The position on the page.

A.  

To measure and monitor service quality within a communications channel.

B.  

To track the time spent by employees working productively.

C.  

To track the number of responses that employees get to their outbound communications.

D.  

To measure the quality of the output of the model.

A.  

eval.enable_ocr

B.  

eval.redo_ocr

C.  

eval.enable_qpu

D.  

eval.use_cuda

A.  

Hybrid and manual data extraction.

B.  

Rule-based and model-based data extraction.

C.  

Rule-based and hybrid data extraction.

D.  

Manual and model-based data extraction.

A.  

Based on the text language and the number of pages.

B.  

Based on the document structure and data points to be extracted.

C.  

Based on the document type and length.

D.  

Based on the number of fields to be extracted.

A.  

Verbose

B.  

Warning

C.  

Information

D.  

Error

A.  

If it does not have any pipeline data.

B.  

If it does not have any running pipelines.

C.  

If it does not have any deployed packages.

D.  

If it does not have any scheduled pipelines.

A.  

Scanned letters

B.  

Shared email inboxes

C.  

Call data in real-time

D.  

Real-time chat data

A.  

It allows access from outside of the UiPath environment.

B.  

It provides additional security measures for the ML Skill.

C.  

It enables automatic updates and enhancements to the ML Skill without user intervention.

D.  

It allows access from inside of the UiPath environment.

A.  

5-10 document samples from each class.

B.  

10-20 document samples from each class.

C.  

20-50 document samples from each class.

D.  

50-200 document samples from each class.

A.  

The understanding of the parent-label relationship when assigning label predictions.

B.  

The difference between label suggestions and label predictions in a training process.

C.  

The identification of a specific span of text as a value for a particular entity type.

D.  

The model's confidence that a specific concept exists within a communication.

A.  

They were trained using the "Search" option extensively.

B.  

They were trained using the "Shuffle" option extensively.

C.  

They have low recall.

D.  

They lack training examples.

A.  

Project Manager

B.  

Solution Architect

C.  

Automation Developer

D.  

Business Analyst

A.  

Out-of-the-box models are custom-built models created specifically for each project.

B.  

Out-of-the-box models are pre-trained models that cover a wide range of document types.

C.  

Out-of-the-box models require extensive training and customization before they can be used effectively.

D.  

Out-of-the-box models are only available for specific languages and cannot be adapted for multilingual documents.

A.  

Install UiPath.IntelligentOCR.Activities package.

B.  

Install a newer version of Studio.

C.  

Install UiPath.Documentunderstanding.ML.Activities package.

D.  

Nothing to do, it should be already there.

A.  

String.Concat(UserNames,",")

B.  

String.Concat(",", UserNames)

C.  

String.Join(UserNames,",")

D.  

String.Join(",", UserNames)

A.  

UiPath.IntelligentOCR.Activities.TaxonomyManagement.LoadTaxonomy

B.  

UiPath.IntelligentOCR.Activities.TaxonomyManagement.LoadFile

C.  

UiPath.IntelligentOCR.Activities.TaxonomyManagement.LoadTaxonomyFile

D.  

UiPath.IntelligentOCR.Activities.TaxonomyManagement.LoadFileTaxonomy

A.  

Inspecting data at a certain point during debugging by evaluating variables, arguments, or statements.

B.  

Modifying the values of variables and arguments at runtime during the debugging process.

C.  

Displaying the next activity to be executed and its parent containers when the project is paused in debugging.

D.  

Displaying the performance analysis of all the operations, showing the execution time of each activity.

A.  

Pre-trained. fine-tunable, and reviewed.

B.  

Pre-trained. custom training, and fine-tunable.

C.  

Custom training, fine-tunable, and reviewed.

D.  

Pre-trained, custom training, and reviewed.

A.  

An area of machine learning concerned with artificial neural networks. These are a series of algorithms that aim to recognize relationships in a set of data through a process that mimics biological neural networks.

B.  

The theory and development of computer systems that are able to perform tasks that normally require human intelligence and decision making.

C.  

A sub-field of artificial intelligence that enables systems to learn from data. Systems learn from previous experience and information to deduce and predict future information. To do this they use algorithms that learn to perform a specific task without being explicitly programmed.

D.  

A branch of artificial intelligence that deals with analyzing, understanding, and generating human natural languages. For example, NLP enables computers to hear speech, read text, interpret the text/speech or measure the sentiment.

A.  

Model-based data extraction.

B.  

Hybrid data extraction.

C.  

Manual data extraction.

D.  

Rule-based data extraction.

A.  

Existence of a non-empty root folder, requirements.txt file, and train.py file in the root folder which implements a class Main. The class is further validated to implement an__init_function.

B.  

Existence of a non-empty root folder, requlrements.txt file, and main.py file in the root folder which implements a class Main. The class is further validated to implement an__init and a predict function.

C.  

Existence of a non-empty root folder, main.py file in the root folder which implements a dass Main. The class is further validated to implement an__init__and a predict function.

D.  

Existence of a requirements.txt file, and main.py file which implements a class Main. The class is further validated lo implement an__init__and a predict function.

A.  

Out-of-the-box models from UiPath and UiPath technology partners.

B.  

Out-of-the-box models from UiPath technology partners and custom models.

C.  

Out-of-the-box models from UiPath, UiPath technology partners, open-source models from the community, and custom models.

D.  

Out-of-the-box models from UiPath technology partners, and open-source models from the community.

A.  

Variables can only be used within the main workflow, while arguments can only be used within invoked workflows.

B.  

Variables are used for input data, while arguments are used for output data.

C.  

Variables have a limited scope within a specific workflow or sequence, while arguments are used to pass data between workflows.

D.  

Variables are global and arguments are local.

A.  

To generate custom data fields for extraction.

B.  

To optimize extraction results and automatically validate them.

C.  

To create complex data manipulation operations.

D.  

To handle exceptions during the extraction process.

A.  

The OCR engine is used to extract text from the image document.

B.  

The extracted data is exported as a dataset.

C.  

The target fields are extracted from the document and sent to Action Center for human validation.

D.  

The documents are included in one of the taxonomy document types or skipped.

A.  

Name. Data Type. Attribute name, and Color.

B.  

Name, Data Type. Attribute name. Shortcut, and Color.

C.  

Name, Shortcut, and Color.

D.  

Name. Input to be labeled. Attribute name. Shortcut, and Color.

A.  

Mean Average Precision.

B.  

Number of labels.

C.  

Volume of data uploaded.

D.  

Number of annotations.

A.  

*.doc*

B.  

*.doc, *docx

C.  

*.doc

D.  

Microsoft Word

A.  

RPA Developer.

B.  

Process Controller.

C.  

Administrator.

D.  

Data Scientist.

A.  

Document Manager.

B.  

Digitization.

C.  

Data Manager.

D.  

ML Extractor.

A.  

Document Types. Taxonomy. Languages.Success Criteria. OCR, ML Model.

B.  

Document Types. Taxonomy. Templates. Languages. Scans/Digital docs, Success Criteria. Technical Criteria.

C.  

Document Types. Technical Criteria, Templates. Languages, Scans/Digital docs, OCR quality.

D.  

Document Types. Templates. Technical Criteria. Languages. Scans/Digital docs. Success Criteria. OCR engines.

A.  

The Document Understanding Process can't use Queues because items waiting for Human Validation for more than 10 days will be marked as Abandoned.

B.  

The Document Understanding Process can use Queues but the Auto Retry Functionality should be disabled.

C.  

The Document Understanding Process can use Queues but the Auto Retry Functionality should be enabled.

D.  

The Document Understanding Process can't use Queues because items waiting for Human Validation for more than 24h will be marked as Abandoned.