Yext site search takes in unstructured data
Yext is upgrading its Answers platform for Web sites with the "Orion" update to its search algorithm due out on March 17. This will add advanced natural language processing (NLP) to answer complex questions against unstructured, long-form documents.
The extractive QA algorithm adds a powerful dimension to Yext’s site search solution. For example, when someone asks a specific question, such as a customer searching a retailer’s site for how to assemble one of their products, extractive QA scans a business’s webpages, blog posts, help articles, or product manuals to find the most relevant word, sentence, or paragraph, then delivers a direct answer in the form of a rich snippet at the top of the results page.
This is distinct from traditional keyword search, which relies on only a single algorithm to match the query’s keywords with the document that contains the most mentions of the keywords, then delivers the end user a list of links to click through.
“At its core, modern search is about understanding human language and delivering a relevant, direct answer. Yet most websites are still stuck in the early 2000s, using basic keyword search technology that returns a list of blue links,” said Marc Ferrentino, Chief Strategy Officer at Yext.
“At Yext, we’re building advanced, multi-algorithm answers search so that businesses and organizations can deliver world class experiences that help grow their business. The extractive QA algorithm update to our platform gives our customers an even bigger competitive edge and further helps accelerate their digital transformations.”
With extractive QA, Yext now offers a comprehensive multi-algorithm search platform that allows businesses to leverage structured, semi-structured, and unstructured data to deliver a seamless, answers-forward search experience to every customer query.