Smarter Enterprise Search using Amazon Kendra

Overview

Often your employees and customers have been trying to find the right information, only to end up clicking unrelated links for hours and can’t find the data which decreases productivity and creates inefficiencies. Additionally, implementing an enterprise search service within a large organization can be difficult. Today most data are unstructured and siled, keyword searches lack context, and search results are often inaccurate. Amazon Kendra solves this using power of machine learning. Amazon Kendra uses natural language understanding and reading comprehension to deeply understand the meaning of the content and the question.

Introduction to Amazon Kendra

Amazon Kendra is an intelligent search service based on machine learning (ML). Amazon Kendra transforms search for your websites and apps so your employees and customers can easily find the content they’re looking for, even if it’s scattered across multiple locations and content repositories across your organization.

Custom Document Enrichment allows you to call external services such as Amazon Comprehend, Amazon Textract, and Amazon Transcribe to extract text from images, transcribe audio, and analyze video. For more information about using custom document enrichment, see Content and Metadata Enrichment to Enhance Your Search Experience with Custom Document Enrichment in Amazon Kendra. With the new Custom Document Enrichment feature in Amazon Kendra, you can now pre-process your documents during ingest and enrich your documents with new metadata.

Amazon Kendra supports various document formats such as Microsoft Word, PDF, and text. Working with a leading Edtech customer, we were asked to create an enterprise search solution that also uses images and PPT files. This post focuses on extending document support in Amazon Kendra so that you can pre-process text images and scanned documents (JPEG, PNG, or PDF) to be searchable. The solution combines Amazon Textract for document pre-processing and Optical Character Recognition (OCR) and Amazon Kendra for intelligent search.

How Amazon Kendra works?

Amazon Kendra provides search functionality to your application. It indexes your documents directly or from your third-party document repository and intelligently provides relevant information to your users. You can use Amazon Kendra to create an updatable index of documents of various types. For a list of document types supported by Amazon Kendra, see Document Types. Amazon Kendra integrates with other services. For example, you can power Amazon Lex chatbots with Amazon Kendra search to provide helpful answers to user questions. You can use an Amazon Simple Storage Service bucket as a data source for Amazon Kendra to connect and index your documents. And you can set access policies or resource permissions using AWS Identity and Access Management.

Amazon Kendra has the following components:

• An index that contains your documents and allows you to search them.
• The data source that stores your documents and that Amazon Kendra connects to. You can automatically synchronize your data source with your Amazon Kendra index, so your index stays up-to-date with your source repository.
• A Document Add API that adds documents directly to the index.

Benefits of Amazon Kendra

• Access to the Content: With Amazon Kendra, we can easily access content from various repositories such as SharePoint, Amazon S3, ServiceNow, and Salesforce into a centralized index that allows you to search all questions from your data to find the exact answer.
• User access control: Within a few clicks, we can configure the index, connect the relevant data sources, and start using Kendra to find answers to our questions.
• Fine-tune search results: We can refine the search results by manually adjusting the importance of data sources or by using our tags.
• Connectivity: Amazon Kendra can connect to third-party data stores or data sources, such as Microsoft SharePoint. You can easily index and search documents using a data source.

Conclusion

As you can see, setting up Amazon Kendra is very easy. So, you can integrate Kendra with your app by using Kendra’s API. Amazon Kendra could be the tech tool that proves how innovation is evolving enough that most industries will find it worthwhile. The company just needs to connect its data stores to the search engine to get the engine to work finding the answers people need to know.

About CloudThat