Predictive AI Basics Quiz
31 Questions
0 Views

Choose a study mode

Play Quiz
Study Flashcards
Spaced Repetition
Chat to lesson

Podcast

Play an AI-generated podcast conversation about this lesson

Questions and Answers

True or False. Predictive AI analyzes data to predict patterns and outcomes.

True

True or False. Generative AI uses machine learning to analyze data sets, finding patterns and making relationships within the data in order to generate original, similar content.

True

Select all that apply. What are some common reasons companies are hesitant to use AI?

  • Data privacy (correct)
  • Response accuracy (correct)
  • Increases headcount
  • Job displacement (correct)
  • You are on a website asking questions to an AI-powered chatbot, which type of AI is being used?

    <p>Generative AI</p> Signup and view all the answers

    You are receiving hundreds of documents that vary from invoices, purchase orders, contracts and general inquiries. Which type of AI can help identify the documents so they can be routed to the appropriate department?

    <p>Predictive AI</p> Signup and view all the answers

    True or False. AI skills enable a developer to build, configure, and train a custom AI model to meet a business use case without the need for any code.

    <p>True</p> Signup and view all the answers

    True or False. AI Copilot is an assistant that gives the developer or user a “head start,” but it doesn’t accomplish tasks independently.

    <p>True</p> Signup and view all the answers

    You received a document that is several pages long. You want to get a short summary of its content, instead of spending time reading through the entire document. Which one of Appian’s AI capabilities would you use?

    <p>AI Skills - Document Summarization</p> Signup and view all the answers

    You want to digitize a PDF form and create a functional Appian interface. Which one of Appian’s AI capabilities would you use?

    <p>AI Copilot - PDF to Interface</p> Signup and view all the answers

    Select three. Which of the following are benefits of using AI at Appian?

    <p>Process Automation</p> Signup and view all the answers

    What are AI skills used for?

    <p>Automate common business tasks such as classification, extraction, summarization, or generation</p> Signup and view all the answers

    Select three. Which of the following are reasons to use AI skills?

    <p>They are private and do not rely on any third-party services</p> Signup and view all the answers

    Select three. Which three AI skills use predictive AI?

    <p>Structured document extraction</p> Signup and view all the answers

    Select three. Which of the following processes can be automated using AI skills?

    <p>Identifying PII in emails</p> Signup and view all the answers

    Acme Insurance wants to create an AI skill that compares the text of an incoming insurance claim to the text of an insurance policy to determine whether or not the claim adheres to the policy, but they don’t see their use case listed. What should they try next?

    <p>Create a prompt using Prompt Builder</p> Signup and view all the answers

    You are creating an AI skill to classify emails for a technical support department. This department typically receives the following types of emails: technical issues and troubleshooting, incident reports, and user feedback and suggestions. How many email types should you create and upload samples for?

    <p>Three: There should be an email type for every category of email the department expects to encounter in production.</p> Signup and view all the answers

    True or false: In order to test your model, you have to first publish your model, and then add and configure a smart service in your process model.

    <p>False</p> Signup and view all the answers

    Which of the following are characteristics of a representative dataset?

    <p>Gathered in roughly equal numbers and Properly labeled</p> Signup and view all the answers

    An important part of document extraction is ________, which helps the model understand what the data looks like and where it might appear in future documents.

    <p>Labeling</p> Signup and view all the answers

    Select two. If your documents contain ________ or ________, you will need to complete a reconciliation task by adding a Reconcile Doc Extraction smart service to your process model.

    <p>Checkboxes</p> Signup and view all the answers

    True or false: The reconciliation task is designed to retrain the model to help Appian learn to map document data to the correct fields.

    <p>False</p> Signup and view all the answers

    Match each term with its definition.

    <p>Accuracy = The ratio of total correct predictions (true positives and negatives) to total predictions Precision = A measure that indicates how many are actually positive out of those predicted positive Recall = A measure that indicates how many were predicted positive out of all real positive case F-1 Score = A metric that takes into account both precision and recall</p> Signup and view all the answers

    Match each term with its definition.

    <p>Accuracy = The ratio of total correct predictions (true positives and negatives) to total predictions Precision = A measure that indicates how many are actually positive out of those predicted positive Confidence Score = A metric that indicates the likelihood that the model predictions are correct Recall = A measure that indicates how many were predicted positive out of all real positive cases</p> Signup and view all the answers

    A model classified 150 documents in total. The model classified 70 as purchase orders. Out of the 70 classified as purchase orders, 60 were actually purchase orders. The model’s precision is ______.

    <p>0.86</p> Signup and view all the answers

    You want to extract key information, such as dates and line item details, from invoices. The accuracy of the extracted information is important, even if it means missing some details, as inaccuracies could lead to accounting errors that would require manual correction. Which metric should you pay the most attention to ensure the model meets your requirements?

    <p>Precision</p> Signup and view all the answers

    What metric should you pay more attention to when you are working with an imbalanced data set?

    <p>F-1 Score</p> Signup and view all the answers

    A team’s model finished training and is ready to be assessed. There is a higher cost if their model receives a W-2 form that contains a SSN, fails to classify the form as a W-2, and instead misclassifies it as a job interest form. Which metric should they care most about during their assessment?

    <p>Recall</p> Signup and view all the answers

    When should you retrain a model? Select all that apply.

    <p>Retrain periodically in case your data changes.</p> Signup and view all the answers

    What changes when you adjust a model’s temperature?

    <p>Creativity</p> Signup and view all the answers

    You created an AI skill for summarizing documents. How do you use the AI skill in a process?

    <p>Add and configure the Execute Generative AI Skill smart service</p> Signup and view all the answers

    Select three. Which of the following should you include when building a generative AI prompt?

    <p>Specific use case</p> Signup and view all the answers

    More Like This

    Use Quizgecko on...
    Browser
    Browser