Amazon Web Services announced the general availability of its intelligent enterprise search tool to help businesses overcome the complexity of modern data architectures. Amazon Kendra enables organizations to index information across multiple data silos and ensure it is easy to search and access by those with the necessary permissions. The service uses machine learning (ML) techniques to allow users to provide queries in natural language, rather than relying on narrow keyword searches that are unlikely to produce the most relevant or accurate results.
Amazon Kendra
The ability to efficiently search the vast amounts of information stored in internal data storage has eluded businesses for many years, as the troves of unstructured data (stored in many formats, in multiple storage locations) are underappreciated. used AWS believes Kendra is responding to this persistent problem by organizing corporate data so employees can leverage all of their company's data. Swami Sivasubramanian, vice president of Amazon Machine Learning at AWS, says enterprises have a lot to gain by addressing the chaotic and random nature of data architectures. "Our customers often tell us that research in their organization is difficult to perform, slows down productivity, and often doesn't work because their data is scattered across many silos in many formats," he said. "The use of keywords is also counterintuitive, and the results returned often require browsing through many irrelevant links and documents to find useful information." By relying on ML algorithms to gain contextual understanding, Amazon Kendra allows users to query databases in a more natural way and receive accurate results without having to wade through information overload. The tool currently supports niche language related to industries such as IT, healthcare, finance, law, telecommunications, and more. Deploying Amazon Kendra would not require any ML knowledge and the tool can be configured from the AWS Management Console. It is currently available in the Eastern US, Western US, and Western EU, with other regions to follow soon.