Amazon Kendra is an intelligent search service powered by machine learning. Kendra reimagines enterprise search for your websites and applications so your employees and customers can easily find the content they are looking for, even when it’s scattered across multiple locations and content repositories within your organization.
Amazon Kendra uses machine learning to deliver more relevant answers from unstructured data. Search for general keywords like "health benefits" or ask natural language questions like "how long is maternity leave?” and Kendra will use reading comprehension to give specific answers like "14 weeks”. For more general questions like "how do I configure my VPN?" Kendra gives descriptive answers by extracting the most relevant text passage.
Amazon Kendra also supports FAQ matching and extracts answers from curated FAQs using a specialized model that pinpoints the closest question in the FAQ and returns the corresponding answer.
To complement extracted answers and FAQ matching, Amazon Kendra uses a deep learning semantic search model for accurate document ranking. Overall, this provides a richer search experience, presenting users with specific answers, as well as related content to explore should they need more information.
Amazon Kendra uses machine learning to continuously optimize search results based on end-user search patterns and feedback. For example, when users search “How do I change my health benefits?", multiple HR benefit documents will compete for a top spot. To determine the most relevant document for this question, Amazon Kendra will learn from the user interactions and feedback to promote preferred documents to the top of the list. Amazon Kendra applies incremental learning techniques automatically without the need for machine learning expertise.
Tuning and accuracy
In addition to providing leading-edge accuracy out-of-the-box, customers can fine-tune search results and boost specific answers and documents in the results based on specific business objectives. For example, relevance tuning lets you boost results based on more authoritative data sources, authors, or document freshness. Learn more in our relevance tuning blog post.
To extend Amazon Kendra’s understading of your specific business vocabulary, you can provide your own custom synonyms. Amazon Kendra uses these to automatically expand queries to include content and answers that match the extended vocabulary. For example, when an end-user asks the question “What is a HSA?” Amazon Kendra would return documents that reference ‘Health Savings Account’ or ‘HSA’.
Using connectors is quick and easy, you just add data sources to your Amazon Kendra index and select the connector type. Connectors can be scheduled to automatically sync your index with your data source, so you're always securely searching through the most up to date content. Amazon Kendra offers native connectors for popular data sources like, S3, SharePoint, Salesforce, ServiceNow, Google Drive, Confluence, and many more. In the event a native connector not available, Amazon Kendra offers a custom data source connector as well as a host of partner supported connectors. For more information about Amazon Kendra connector availability, visit the Amazon Kendra connector library.
Kendra uses deep learning models to understand natural language queries and document content and structures for a wide range of internal use cases like HR, operations, support, and R&D. Kendra is also optimized to understand complex language from domains like IT, financial services, insurance, pharmaceuticals, industrial manufacturing, oil and gas, legal, media and entertainment, travel and hospitality, health, HR, news, telecommunications, mining, food and beverage, and automotive. For example, a user searching for HR answers could enter "deadline for filing HSA form" and Kendra would also search for "deadline for filing health savings account form" for broader coverage to get the most accurate answer.
Amazon Kendra includes the functionality to auto-complete an end user’s search query. Query auto-completion not only helps the user reduce typing by about 25%, it also helps them by guiding them towards more precise and commonly asked questions. More precise questions typically result in more relevant and useful answers. For example, if you start typing "Where is" in the search box, Kendra can suggest options like "Where is the IT desk?", or "Where is the cafeteria?" and other related commonly asked questions, to complete the query.
Analytics and ongoing improvement coming soon
In order to provide a continuously improving search experience, Amazon Kendra captures the activities during searches (clicks, thumbs up or down) and will surface metrics and insights to you so that you can take action to make improvements. Kendra provides basic operational metrics like top queries, top documents, and queries per day. Kendra will also provide common quality metrics such as Mean Reciprocal Rank (MRR) and explicit feedback, i.e. thumbs up or down counts. You can use this information to create more relevant FAQs, boost certain data sources with more authoritative content, or train your customer support teams on commonly asked questions.