Enterprise Search Strengthened by Machine Learning

CIOReviewIndia Team | Wednesday, 04 December 2019, 13:37 IST

Enterprise Search Strengthened by Machine LearningIn a bid to enable its enterprise customers gain better insight into business data, Amazon has launched two artificial intelligence services aimed at helping businesses make sense of business data spread across segments and processes. Usage of Machine Learning has been kept on a pedestal while designing these services. The two services - Contact Lens for Amazon Connect and Amazon Kendra are available for its customers using AWS cloud. The services promise to change how enterprise search is done by incorporating new technological capabilities like Natural Language Processing.

Amazon Kendra is a plug and play service and does not require expertise in machine learning for the business user. The service will act as an AI-enabled search engine for Amazon’s enterprise customers. “We think this is a step-change in the capability for internal search,” said Matt Wood, AWS's Vice President of Artificial Intelligence, at an AWS conference in Las Vegas. "We've certainly experimented with different technologies within Amazon, and none of them work very well." 

Equipped with this capability, organizations will now be able to scour data irrespective of the format of the location within enterprise storage. Insightful patterns can be identified by organizations for pre-emptive as well as reactionary decision making. Kendra can be set up from AWS Console. Using machine learning, its builds a search index as per the organization’s documents and FAQs. It applies natural language processing to gauge the intent and context the document carries while looking for relationship between content and documents. Kendra has been designed to be cognitive and is capable of continuously refining its search results based on the past feedback and searches. From the AWS console, one can test and refine the queries. The service improves with time by incorporating user feedback.

 “There is no machine-learning expertise required for either of these services. They are just plug and play. You don’t have to get into all the weeds and get the training data and label the data and all those sorts of things,” said Matt Wood. The AI-enabled enterprise search tool can be implemented on websites as well as interfaces such as chatbots. “As people use this service more, the machine-learning models under the hood get better at ranking and providing relevant answers to the questions,” added Wood. 

Don't Miss ( 1-5 of 25 )