Specify the search range in the Box from Kendra when using the Amazon Kendra Box Connector

Yes, this is possible.

When using the Amazon Kendra Box Connector, you can specify which folders and files to include or exclude in your search. This can be useful for complying with confidentiality requirements by ensuring that only specific documents are indexed and searchable.

I have provided some instruction on how to specify folders/files to be searchable with Kendra Box Connector:

1. Use Inclusion and Exclusion Patterns

Amazon Kendra allows you to define inclusion and exclusion patterns using regular expressions. These patterns can be used to specify which files and folders should be indexed.

1. Create or Update the Data Source:

• When setting up your box data source in the console, you can specify inclusion and exclusion patterns as part of the configuration.

2. Define Inclusion Patterns:

• Inclusion patterns determine which files and folders should be indexed. For example, if you want to include all files in a specific folder, you can use a pattern such as /folder1/.*.

3. Define Exclusion Patterns:

• Exclusion patterns determine which files and folders should be excluded from indexing. For example, to exclude files in a folder named confidential, you can use a pattern like /confidential/.*.

4. Apply the Patterns:

• Apply these patterns in the data source configuration under the Inclusion Patterns and Exclusion Patterns sections.

Example:

 { 
   "DataSourceConfiguration": { 
       "BoxConfiguration": { 
           "BoxFieldMappings": [...], 
            "InclusionPatterns": [ 
               "/projectA/.*", 
               "/projectB/.*"
            ], 
            "ExclusionPatterns": [ 
                "/projectA/confidential/.*", 
                "/projectB/confidential/.*" 
            ] 
         } 
      }     
  }

2. Use Access Control Lists (ACLs)

In addition to using patterns, you can leverage Box’s Access Control Lists (ACLs) to control which files and folders are accessible to Kendra. Ensure that Kendra’s Box account has the appropriate permissions to only the folders and files you want it to index.

3. Customizing Sync Settings

When configuring the Kendra Box Connector, you have the option to customize the sync settings. You can specify:

• Specific folder paths to be included in the sync.
• Filters based on file types, metadata, and other attributes.

1. Navigate to the Kendra Console:

• Go to the Amazon Kendra console and navigate to the Data Sources section.

2. Add or Edit Data Source:

• Add a new data source or edit an existing one.

3. Select Box as the Data Source:

• Choose Box as the data source type.

4. Specify Inclusion and Exclusion Patterns:

• In the data source configuration, add your inclusion and exclusion patterns.

5. Set Permissions:

• Ensure the Kendra service account used has the necessary permissions for the specified folders and files in Box.

Example Configuration in AWS Management Console:

1. Inclusion Pattern Example:

• To include only files in the /public folder, you might add an inclusion pattern like /public/.*.

2. Exclusion Pattern Example:

• To exclude all files in any private folder, add an exclusion pattern like /private/.*.

Configuring these settings, allows you to control the parts of your Box repository that are indexed by Amazon Kendra, this ensures that sensitive or confidential information remains secure and unindexed. You can use this to filter the search capabilities to meet your specific needs and compliance requirements.

For more detail on this topic you can refer to the following sources: https://docs.aws.amazon.com/kendra/latest/dg/data-source-box.html https://aws.amazon.com/blogs/machine-learning/getting-started-with-the-amazon-kendra-box-connector/ https://docs.aws.amazon.com/kendra/latest/dg/data-source-box.html