AWS re:Invent is currently underway in Las Vegas, and the firm’s annual enterprise-focused event has seen a flurry of announcements being made. One of the more intriguing is Amazon Kendra, which is the firm’s new enterprise search tool which is powered by machine learning.
For those developers or enterprises concerned that the addition of machine learning means some knowledge or expertise is required to make the most of the solution, AWS explains that this is quite the contrary.
This as Kendra features pre-built connectors that can be assigned to whatever content repositories your business favours, such as OneDrive or Salesforce. From there users can simply confirm their AWS credentials and begin putting the new enterprise search tool to work.
“Amazon Kendra’s ability to understand natural language questions is at the core of its search engine, so end users have the ability to search for general keywords like ‘health benefits’ or more specific natural language questions like ‘how long is maternity leave?’. Kendra will return specific answers like ’14 weeks’, or for the more general searches Kendra will return the most relevant passage and related documents. Natural language enables you to get more specific answers from anywhere in your data,” AWS notes regarding the natural language understanding of Kendra.
AWS also explains that Kendra is extremely simple to interact with from a customer or user perspective, with basic feedback like a smiley or sad face capable of being understood by the machine learning running things in the background.
The firm has additional features on the way too, with incremental learning, query autocomplete and improved analytics all forthcoming.
As such it looks as if Kendra will be the enterprise equivalent to Alexa, although the company did not state whether that level of intuitive interface is in the offing moving forward.
Also unclear is whether other regions will receive access to Amazon Kendra, with Alexa not supported in certain countries like South Africa for example. That may change once AWS’ data centres spin up locally next year though.
