UiPath UIPATH-SAIV1 Online Practice Questions and Exam Preparation

UiPath UIPATH-SAIV1 Online Questions & Answers

• Question 111:

  For what purpose can data coming from human validation be used in AI Center?

  A. To evaluate the ML model.
  B. To retrain the ML model.
  C. To keep track of corrections done on the extracted data.
  D. To keep track of the number of documents that require human validation.
  B. To retrain the ML model.

  ---

  Explanation/Reference:

  Using Data Labeling with human in the loop Data Labeling enables you to upload raw data, annotate text data in the labeling tool (for classification or entity recognition), and use the labeled data to train ML models. Apart from this, you can use data labeling for human validation on model outputs.

  A common scenario is when you train an extractor or classifier model. When the model prediction falls below a set confidence threshold, that data can be sent to Action Center for human validation. The validated data can be used to retrain the model in order to improve confidence on subsequent model predictions.

  https://docs.uipath.com/ai-center/automation-cloud/latest/user-guide/using-data-labeling-with-human-in-the-loop

• Question 112:

  When adding a new label to a mature taxonomy in UiPath Communications Mining, why is it important to apply it to all of the relevant examples in the previously reviewed set?

  A. Training data for each label should always be evenly spread across the time period covered in the dataset.
  B. If the new label is not applied to relevant previously reviewed verbatims, the model will start predicting other labels less accurately.
  C. It will reduce bias for the new label.
  D. The previously reviewed examples where the new label should also have been applied will cause confusion for the model if not corrected.
  D. The previously reviewed examples where the new label should also have been applied will cause confusion for the model if not corrected.

• Question 113:

  What is the definition of a UiPath Communications Mining data source?

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

  ---

  Explanation/Reference:

  A 'source' refers to a raw collection of messages which can grow over time. For example, a source could be all the responses collected to a survey, the emails in a team mailbox, the transcripts in a messaging channel, or all of the calls against a telephone number.

  Sources are added to datasets in order to build a model to interpret and structure the messages within them.

  Each source can be added to up to 10 different datasets.

• Question 114:

  When is it recommended to use Main-ActionCenter in the context of the Document Understanding Process?

  A. When implementing an attended process.
  B. When testing locally or implementing an attended process.
  C. When testing locally.
  D. When testing locally or implementing an unattended process.
  D. When testing locally or implementing an unattended process.

• Question 115:

  What is the primary purpose of assigning labels to verbatims within a dataset in UiPath Communications Mining?

  A. To apply labels to a maximum amount of data to enhance model performance.
  B. To provide the platform with a collection of keywords associated with each label for predicting other instances.
  C. To formulate a comprehensive training set that closely reflects the entirety of the dataset.
  D. To allocate 20 training examples for each label.
  C. To formulate a comprehensive training set that closely reflects the entirety of the dataset.

• Question 116:

  When training a UiPath Communications Mining model, what is the recommended number of communications to be tagged using the Search feature?

  A. Minimum 10% per label.
  B. Maximum 12 per label.
  C. Minimum 12 per label.
  D. Maximum 10% per label.
  C. Minimum 12 per label.

• Question 117:

  Which of the following use cases is best suited for label sentiment analysis in UiPath Communications Mining?

  A. Analyzing customer survey responses.
  B. Analyzing process inefficiencies in a shared email inbox.
  C. End-to-end automation of requests from inbound workflow tickets.
  D. Automatically routing inbound emails in a shared email inbox.
  A. Analyzing customer survey responses.

• Question 118:

  What type of data extraction methodology should be used, according to best practices, for a fixed layout document (e.g. a form)?

  A. Model-based data extraction.
  B. Hybrid data extraction.
  C. Manual data extraction.
  D. Rule-based data extraction.
  D. Rule-based data extraction.

• Question 119:

  Which UiPath Communications Mining model performance factor assesses the proportion of the entire dataset that has informative label predictions?

  A. Average label performance.
  B. Coverage.
  C. Balance.
  D. Underperforming labels.
  B. Coverage.

  ---

  Explanation/Reference:

  Coverage is a term frequently used in Machine Learning and relates to how well a model 'covers' the data it's used to analyse. In Communications Mining, this relates to the proportion of messages in the dataset that have informative label predictions, and is presented in Validation as a percentage score.

  'Informative labels' are those labels that the platform understands to be useful as standalone labels, by looking at how frequently they're assigned with other labels. Labels that are always assigned with another label, e.g. parent labels that are never assigned on their own or 'Urgent' if it's always assigned with another label, are down-weighted when the score is calculated.

  The visual below gives an indication of what low coverage versus high coverage would look like across an entire dataset. Imagine the shaded circles are messages that have informative label predictions.

• Question 120:

  What is important regarding entities when working with UiPath Communications Mining and training using Clusters?

  A. If there are no entity predictions for a cluster, skip that cluster.
  B. Accept or dismiss entity predictions or manually assign entities that are not predicted.
  C. Entity predictions are never made for clusters.
  D. Training entities is not recommended during this phase.
  B. Accept or dismiss entity predictions or manually assign entities that are not predicted.