![]() ![]() More details can be found in Information sources. For example, you may choose an internal SharePoint site or OneNote as a source for generative answers. In addition to URL considerations, you'll also need to consider authentication for your sources (should there be any). Custom data (internal or external): supply your own content from any source for instance, from a Power Automate Flow, Skill, or other source.These sources can be used by generative answers: Your bot's ability to answer user questions can also use other web sites, external or internal web sources, and knowledge sources such as SharePoint or OneDrive for Business.ĭetails and examples on how you can expand your bot's ability to use generative answers can be found in Generative Answers with Search and Summarize. While generative answers serves as a fallback in this situation, it isn't limited to fallback scenarios. This document is focused on getting you started using generative answers to augment your bot's ability to help customers. This behavior is called "Generative Answers for fallback".įinally, if the user's intent isn't matched by topics or generative answers, the Fallback system topic is called. If a matching intent isn't found in the topics, the bot can use generative answers to attempt answering the query. When a user sends an input to a bot, the bot first looks for topics to run which match the intent of the user prompt. This workflow could take a while and some specialized knowledge - but with generative answers enabled you're up and running from day one. You create individual topics for the most important or most often-asked questions from your customers (which you might have developed based on analytics from previous bots or existing support issues). You test it thoroughly.Īfter testing it, you publish your bot so you can instantly provide answers, help, and guidance to your customers or bot users. You create a bot and enable this capability. Summarize search results into plain language that is delivered to the bot user.Find, collate, and parse relevant information from a specified source (for example, your company's website) or from multiple sources, including Sharepoint and OneDrive for Business.Parse what a user types to determine what they're asking.Now, before involving a live agent, the bot uses natural language processing (NLP) to: Generative answers with search and summarize discusses how your bot can query information sources by introducing generative answers in a node. This document covers how to configure generative answers as a fallback topic when a user intent cannot be addressed by existing bot topics. ![]() If, after two prompts, the bot still couldn't determine the user's intent, the bot escalated to a live agent by using the system Escalate topic. In the past, when a bot couldn't determine a user's intent, it asked the user to rephrase their question. As a result you can quickly create and deploy a functional bot, without having to manually author multiple topics that may not address your customer's questions. Generative answers can be used as primary information sources in your chatbot, or as fallback when authored topics are unable to address a user's query. Generative answers in Microsoft Copilot Studio allow your bot to find and present information from multiple sources (which may be internal or external) without requiring creation of topics. Some articles and screenshots may refer to Power Virtual Agents while we update documentation and training content. Power Virtual Agents capabilities and features are now part of Microsoft Copilot Studio following significant investments in generative AI and enhanced integrations across Microsoft Copilot. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |