Integrate the Amazon Bedrock Knowledge Base with Microsoft SharePoint as a source of information

The Amazon Bedrock Information Base offers base fashions (FMs) and brokers in Amazon Bedrock contextual data out of your firm’s personal sources for retrieval augmentation technology (RAG) to supply extra related, correct, and customised reply. Amazon Bedrock Information Base offers a completely managed RAG expertise.

The sources of data that may function hyperlinks to the information base are continually increasing. This text reveals learn how to use one of many knowledge supply connectors; Microsoft SharePoint, an built-in content material administration and collaboration device that many organizations use to retailer, set up, and share their inside knowledge. For an entire listing of supported knowledge supply connectors, see Knowledge supply connectors.

Resolution overview

Listed here are some associated options for SharePoint sources within the Amazon Bedrock Information Base:

• It offers entry to data saved in SharePoint. The RAG structure queries and retrieves related data from SharePoint sources to supply contextual responses based mostly on person enter.
• It extracts structured knowledge, metadata and different data from paperwork imported into SharePoint to supply related search outcomes based mostly on person queries.
• It offers the power to constantly synchronize incremental SharePoint content material updates.
• It offers supply attributes of responses generated by the FM.

Within the following sections, we stroll via the steps of establishing a information base, configuring sources, and testing the answer.

Stipulations

The next are the conditions required to implement an Amazon Bedrock information base utilizing SharePoint because the connector:

Construct a information base and join sources

Full the next steps to arrange a information base on Amazon Bedrock and hook up with a SharePoint supply:

1. On the Amazon Bedrock console, select information base Within the navigation pane.
2. select Create a information base.

3. exist Information base particulars part, you may select to vary the preset identify and enter an outline of the information base.
4. exist IAM permissions part, choose the IAM function that gives Amazon Bedrock with entry to different AWS providers. You’ll be able to have Amazon Bedrock create a service function or choose a customized function that you’ve got already created.
5. exist Choose knowledge supply half, choose share level.
6. (Non-compulsory) Add tags to your information base. For extra data, see Tag Assets.
7. select Subsequent.

8. exist identify and outline part, you may select to vary the default knowledge supply identify and enter an outline of the information supply.
9. exist supply part, offers the next data:
   1. for Website URLenter the web site URL used to crawl and index RAG content material.
   2. for disciplineenter the area identify related to the information supply. For instance, if the web site URL is https://deloittedasits.sharepoint.com/xyz.aspxthe area worth might be deloittedasits.
   3. beneath superior settingsretain the default choice.

When changing your knowledge to an embed, Amazon Bedrock encrypts your knowledge utilizing keys owned and managed by AWS by default. To make use of your personal AWS Key Administration Service (AWS KMS) key, select Customized encryption settings (superior) and select a key. For extra data, see Encryption of Transient Knowledge Storage Throughout Knowledge Ingestion.

You can even choose a knowledge removing coverage to your knowledge supply from the next choices:

• delete – When deleting a information base or supply useful resource, all underlying knowledge belonging to that supply is deleted from the vector retailer. Word that the vector retailer itself is just not deleted, solely the underlying knowledge. This flag is ignored if the AWS account is deleted.
• Preserve – After deleting a information base or supply useful resource, all underlying knowledge within the vector retailer is retained.

For extra details about managing information bases, see Managing Sources.

10. exist confirm Partially, the supported authentication methodology is ready to OAuth 2.0.
    1. for Tenant IDenter your tenant ID. See the Register a brand new utility within the Microsoft Azure portal part of this text to acquire a tenant ID.
    2. for AWS Secret Supervisor Key, enter AWS Secrets and techniques Supervisor. See the Creating Secrets and techniques Supervisor Keys for SharePoint Sources part of this text to acquire the important thing.

SharePoint sources require credentials to connect with SharePoint On-line websites utilizing the Microsoft Graph API. To do that, create a brand new Secrets and techniques Supervisor key. These credentials usually are not utilized in any entry logs for the SharePoint On-line web site.

11. exist Metadata settings part, you may choose any content material kind to incorporate or exclude.

12. exist Content material chunking and parsing half, choose default.

13. select Subsequent.
14. exist embedded mannequin half, choose Titan Embeddings G1 – Textual content or different acceptable embedding mannequin.
15. exist vector database half, choose Rapidly create new vector shops Create vector storage for embedding.
16. select Subsequent.

17. superior evaluation and create web page, confirm your choice and choose create.

Information base creation needs to be accomplished.

The information base utilizing SharePoint as the information supply has now been established. Nevertheless, synchronized sources are required to crawl web site URLs and index related content material.

18. To start out this course of, choose your knowledge supply on the information base particulars web page and choose synchronous.

Register a brand new utility on the Microsoft Azure portal

On this part, we register a brand new utility within the Microsoft Azure portal. We retrieve the tenant person ID on this step to be used when configuring the information supply for the Amazon Bedrock information base. Full the next steps:

1. Open the Azure portal and log in together with your Microsoft account. If you do not have an account, you may create one or contact your group’s administration workforce.
2. select New registration.
3. Present the next data:
   1. for identifypresent the identify of your utility. We name this app TargetApp. The Amazon Bedrock information base makes use of TargetApp to connect with a SharePoint web site to crawl and index supplies.
   2. for Who can use this utility or entry this API, select Solely accounts on this group’s listing (solely<租戶名稱> – single tenant).
   3. select Register.
   4. Word the appliance (consumer) ID and listing (tenant) ID Overview You may want them later while you want them TargetApp-ClientId and TenantId.
4. select API permissions Within the navigation pane.
5. Configure permissions as follows:
   1. select Add new permissions.
   2. select Microsoft Map.
   3. select Delegate permissions.
   4. select Learn all exist person half.
   5. select Learn all exist Group members half.
   6. select Full management.All exist web site half.
   7. select Add new permissions. This permission permits the appliance to learn knowledge about logged-in customers within the group’s listing.
   8. On the choices menu (three dots), choose Delete permissions.
   9. Delete the unique Studying – Delegate permit.
   10. select Grant admin consent is the default listing.

6. select Certificates and secrets and techniques Within the navigation pane.
   1. select New Buyer Secrets and techniques.
   2. for describeenter an outline, comparable to an outline for my consumer key.
   3. Choose a worth Expired. In manufacturing, you want to manually rotate keys earlier than they expire.
   4. select Add to.
   5. Word the worth of the brand new secret. You will have it later if you find yourself requested for the consumer secret (TargetApp-ClientSecret).
7. Or, choose proprietor Add any further house owners to the appliance. The proprietor will be capable of handle permissions for the Azure AD utility (TargetApp).

Create Secrets and techniques Supervisor keys for SharePoint knowledge sources

Full the next steps to create a Secrets and techniques Supervisor key to connect with the SharePoint on-line web site listed as the positioning URL within the supply:

1. On the Secrets and techniques Supervisor console, choose Save new secret.
2. for secret kindselect one other secret.
3. for key/worth pairenter the next:
   1. username
   2. password
   3. Consumer ID
   4. consumer secrets and techniques
4. for encryption keyselect aws/secretsmanager.
5. select Subsequent.
6. exist Secret identify and outline part, enter a reputation and elective description for the important thing.
7. Add any related tags in Label
8. go away Useful resource permissions and copy secret as default worth.
9. select Subsequent.
10. exist Configuration rotation part, preserve the default values ​​or modify them in line with your group’s insurance policies.
11. select Subsequent.
12. Evaluation your choices and select store.
13. On the Key Particulars web page, observe the important thing ARN worth that you’ll use as the important thing when constructing the repository for Amazon Bedrock.

Take a look at resolution

Full the next steps to check the information base you created:

1. On the Amazon Bedrock console, select information base Within the navigation pane.
2. Choose the information base you created and choose check.

3. Select the appropriate mannequin to check and select Apply.

4. Enter a query about content material in your SharePoint web site.

clear up

When you created a brand new information base to check out this text and don’t plan to make use of it additional, delete the information base in order that prices don’t accrue to your AWS account. For directions, see Managing the Information Base.

in conclusion

On this article, we present you learn how to arrange an Amazon Bedrock information base utilizing SharePoint On-line as a knowledge supply. By connecting SharePoint On-line as a knowledge supply, staff can use pure language to work together with organizational information and knowledge saved in SharePoint to simply discover related data, extract key factors, and achieve beneficial insights. This will considerably enhance productiveness, decision-making and information sharing inside a corporation.

Do that characteristic on the Amazon Bedrock console immediately! Please see the Amazon Bedrock Information Base for extra data.

Concerning the writer

Surendhar Jayaveli It is a gentleman. Options Architect dwelling in Nashville, TN. He’s a passionate expertise fanatic who enjoys working with shoppers and serving to them construct revolutionary options.

Abhi Patrolla It is a gentleman. Options architect based mostly within the New York space, helping shoppers with cloud transformation, AI/ML, and knowledge planning. He’s a technique and expertise chief who advises senior executives and engineers on cloud methods to drive innovation and optimistic impression.