Posts

Showing posts from March, 2021

Power Platform : AI Builder Models

 AI Builder is one of core components of platform and Microsoft has been working hard to make business usages  In the Microsoft's word:  AI Builder is : AI Builder is a Power Platform capability that helps you improve your business performance by automating processes and predicting outcomes. By using AI Builder, you can quickly bring AI to your apps and flows that connect to your business data that is stored in the underlying data platform (Microsoft Dataverse) or in various cloud data sources, such as SharePoint, OneDrive, or Azure. AI capabilities are brought to your apps and flows by models. A model can be built and customized by you, or it can be a prebuilt model that is ready to be used right away. Microsoft provides several pre-built models for business uses:   Form Processing : Form Processing lets you read and save information from standard documents such as Invoices or tax documents. It can use in Power Apps and Power Automates. A Form Processing model will learn from docu

Notes for PL 200 : Manage Topic in Power Virtual Agents

 Bots are being used to expend self- service support experiences for customers and to help ease agent workloads by handling specific types of support topic that are the highest volume and easily automated. Bot Topics: In Power Virtual Agents, Topics represent paths that a customer can be taken in while interacting with a bot.  The Topic that is used and the path that is followed within an individual topic is in response to the data that is entered by customers in the conversion panel. Power Virtual Agents Topics consist of two primary elements: Trigger Phrases: Phrases, keywords, or questions that are entered by users and relate to a specific issue. Conversation Nodes: Define how a bot should respond and what it should do. Entities: Power Virtual Agents conversation use natural language understanding to identify a user's intent based on what they entered into the conversation window. Power Virtual Agents comes with a set of prebuilt entities out of the box. these entities represen

Notes for PL 200 : Receipt Processing in AI Builder

Important points for Exam:   Receipt Processing lets read and save key information from common Sales receipts, such as receipts that are used in restaurants, gas stations, retails and more. The prebuilt AI model uses state- of art optical character recognition (OCR) to extract printed and hand written text receipts. such as merchant details, transaction date and time, a list of purchased items, taxes and total. No training or Prior configuration is required to use receipt processing. It only needs to provide an image or a PDF file and the model will instantly return the results. Only English receipts from the United States are correctly supported. full page receipts, such as hotel receipts or invoices are not supported Common business problems that can be solved with receipt processing include:- Travel and Expense  (T & E) Accounting and Auditing