News

AI-Powered Search Solution Comprehends 'True Intent' of User Inquiries

• By John K. Waters
• 05/05/2021

Artificial intelligence (AI) software maker, Kyndi, today unveiled a new intelligent search platform designed to understands the 'true intent' of a business user's question against documents, emails, manuals, and other forms of textual data to return contextually relevant answers.

The San Mateo, CA-based company is billing its new Cognitive Search Platform as a next-gen tool powered by a proprietary cloud AI platform.

"Unlike traditional AI search tools that aren't intuitive and provide only limited, keyword-based results, the Kyndi platform augments humans by mimicking the human thought process," Ryan Welsh, CEO and founder of Kyndi, said in a statement. "Employees may ask questions of data a dozen ways that all mean the same thing. But Kyndi has the context-driven intelligence to not only understand the intent of a user's question, but also to explain — at a granular level — why an answer was returned. And this is exactly where traditional search techniques fail to deliver the experience that business users are looking for."

Kyndi is probably best known as the creator of one of the first Explainable AI platforms. According to former Carnegie Melon University professor Anupam Datta, explainability is about illuminating how neural networks reach their decisions, so that human users can understand, appropriately trust, and effectively manage AI. Unlike black-box AI search solutions, Kyndi's cognitive search platform provides detailed explainability of its results by providing end users with the exact sentence, page, and document-level location details to deliver trust in the results found.

Hayley Sutherland, senior IDC research analyst focused on conversational AI, points to the potential of solutions like Kyndi's to improve business efficiencies. "Workers struggle to find timely, relevant answers and solutions, even though they and their colleagues are creating more content and information every day," Sutherland said in a statement. "By using AI technologies such as machine learning and natural language processing to build an intelligent knowledge base, organizations can create a single source of truth and a unified view of information, and even begin to make new connections across their data."

The Kyndi Cognitive Search platform employs a neuro-symbolic technique to process language, which it couples with AI algorithms. The platform was designed to interpret industry-specific terminology, acronyms, and synonyms without any data training or AI skills, the company says.  
It can also handle such language variations as semantically equivalent words and misspelled queries. 

Last month, industry analysts at CB Insights included Kyndi in its annual AI 100 ranking, which showcases the100 "most promising" private AI companies in the world.