Amazon SageMaker Canvas, introduced in 2021, allows business analysts to build and deploy machine learning (ML) models without coding. With recent updates, SageMaker Canvas now supports foundation models (FMs), enabling users to query documents from their knowledge base. Integration with Amazon Kendra allows for context-specific and secure retrieval of information. The feature can be easily set up by configuring the Kendra index and enabling the query documents toggle in SageMaker Canvas.
Empower Your Business with Amazon SageMaker Canvas: A No-Code AI Solution
Enterprises are increasingly turning to Machine Learning (ML) to solve complex problems and improve outcomes. However, building and deploying ML models traditionally required deep technical skills. That’s where Amazon SageMaker Canvas comes in. It allows business analysts to build, deploy, and use ML models without writing code.
With the recent expansion of SageMaker Canvas, users can now ask questions and get context-specific responses grounded in their enterprise data. This opens up new use cases for no-code ML, such as formulating responses consistent with an organization’s specific vocabulary and quickly querying lengthy documents.
How It Works
To get started, a cloud administrator configures and populates Amazon Kendra indexes with enterprise data as data sources for SageMaker Canvas. Users can then select the index where their documents are located and start exploring and researching with confidence, knowing that the output is backed by trusted sources-of-truth.
SageMaker Canvas uses state-of-the-art Foundation Models (FMs) from Amazon Bedrock and Amazon SageMaker JumpStart. Conversations can be started with multiple FMs side-by-side, comparing outputs and making generative AI accessible to everyone.
Reducing Hallucinations with Retrieval Augmented Generation (RAG)
Foundation models can sometimes produce generic or incorrect responses. To address this, SageMaker Canvas leverages Retrieval Augmented Generation (RAG) architectures. RAG retrieves data from external sources, such as document repositories or databases, to perform in-context learning and provide accurate answers to user queries.
With the integration of Amazon Kendra, enterprises can enrich the chat experience in Canvas with a powerful knowledge management system. By connecting SageMaker Canvas to knowledge bases in Kendra, organizations can keep their proprietary data secure while benefiting from the natural language capabilities of FMs.
Setting Up Canvas to Query Documents
Setting up Canvas to query documents served through Kendra indexes is a straightforward process. It involves creating a Kendra index, setting up the Kendra Amazon S3 connector, and configuring IAM permissions for Canvas. Once these steps are completed, a cloud administrator can enable the “Query Documents with Amazon Kendra” toggle in the SageMaker Domain settings and select the desired Kendra indices.
Using the Query Documents Feature for Chat
Within SageMaker Canvas, users can access the Query Documents feature from within a chat session. By enabling Query Documents, users can choose from a list of Kendra indices enabled by the cloud administrator. They can then ask questions, and Canvas will retrieve relevant snippets from the selected index to generate accurate responses. Users can also click on source documents to view previews and see the excerpts used by the FM.
Conclusion
Amazon SageMaker Canvas empowers businesses to harness the power of AI without the need for coding skills. By connecting Canvas to knowledge bases in Amazon Kendra, organizations can provide accurate, relevant responses to user questions using their domain knowledge and trusted sources-of-truth. With the recent addition of the Query Documents feature, SageMaker Canvas makes it easy for enterprises to use large language models and their enterprise knowledge to power secure and seamless chat experiences.
To learn more about SageMaker Canvas and how it can help your business leverage AI, check out the SageMaker Canvas announcement. For practical AI solutions and advice on AI KPI management, connect with us at hello@itinai.com or visit our website at itinai.com.
Spotlight on a Practical AI Solution: Consider the AI Sales Bot from itinai.com/aisalesbot. This solution automates customer engagement and manages interactions across all stages of the customer journey. Discover how AI can redefine your sales processes and customer engagement at itinai.com.
