Our investment into Q&A does not stop here. Choose a type which fits the term you’re trying to define (for example, “best” is an adjective), then follow the prompts to define what it means in the context of your data. There, you’ll be able to define a variety of relationships, including verb, adjective, noun, preposition, and more. You can get into the Q&A setup menu using the gear icon on the Q&A visual or the Q&A setup option in the Modeling tab of the ribbon, then selecting the new Relationships tab. There are many types of these linguistic relationships, so we built a new tab entirely to help you create and manage linguistic relationships for your data. Best in the previous example is one asking “who sold the most books” requires us to know that stores sell books, connecting stores to books. The other half of the terms (adjectives, verbs, prepositions, adverbs) can’t be defined with such straightforward mappings because they must be understood as a part of a phrase – they qualify other terms or relate other terms together. However, synonyms (nouns) are only half of the picture. This way, authors can explicitly define console as referring to products, and users will always receive the correct answers when they ask similar questions in the future. To help authors ensure that the Q&A visual provides consistent and accurate answers based on the unique language their report consumers actually use, we introduced Q&A setup tools with an emphasis on providing Q&A with synonyms for column and table names in the model. These terms are contextual, however – users could mean something completely different asking for console and best in other industries, organizations, or even datasets. For example, answering “what are our best consoles this year?” may require connecting the term consoles to the name products in the model, and understanding that the term best corresponds to the highest sales values. However, while the Q&A engine is good at answering precise questions about data, it may not be able to associate every word or phrase a user inputs with data in the model. It offers users a way to explore their data in ways not covered by the rest of the report without requiring deeper knowledge of their data model or report authoring. The Q&A visual is an effective way to help users further understand their data by asking questions and receiving answers in the form of visuals. Edit linguistic relationships in Q&A setup Please continue to submit your feedback directly in the comments of this blog post or in our feedback forum.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |