Amazon Kendra
Managed Connection - Amazon Kendra
Amazon Kendra (https://aws.amazon.com/kendra/) is a highly accurate and easy to use enterprise search service thats powered by machine learning. Kendra delivers powerful natural language search capabilities. It’s also a fully managed Amazon Service.
Please follow the link below to learn more and get started with Amazon Kendra https://www.youtube.com/watch?v=ow5DFGICHoI
Configuration on Amazon Kendra console
(1) Create Index
An index provides search results for documents and frequently asked questions (FAQ) that it has indexed. An index can contain documents that are indexed from a data source, documents that are added directly to the index, and FAQs.
(2) Create Data Source
A data source is a location, such as an Amazon Simple Storage Service (Amazon S3) bucket, where you store the documents for indexing. Data sources can be automatically synchronized with an Amazon Kendra index so that new, updated, or deleted documents in the source repositories are included in searches. Supported data sources are:
Amazon S3 buckets
Amazon RDS for MySQL and Amazon RDS for PostgreSQL databases
Microsoft OneDrive for Business
Microsoft SharePoint Online
Salesforce sites
ServiceNow instances
Supported document formats are plain text, Microsoft Word, Microsoft PowerPoint, HTML, and PDF
VoiceWorx
The VoiceWorx platform allows for the rapid integration of Amazon Kendra to various voice channels such as, Alexa, Lex, Google Assistant and more.
To Configure Amazon Kendra, browse to: Integration > Connect System > Connect External System in the VoiceWorx platform and select options as shown below.
You can get your required AWS Credentials from the AWS IAM console. Visit Video: How to get Amazon (AWS) Key and Secret Make sure the AWS account has the required permissions to access Amazon Kendra.
Once you have saved settings, browse to: Custom Skills > Manage Skills and create a new Custom Skill by clicking on “Create New Custom Skill” button.
Now, create a Data Source for the Amazon Kendra by navigating to Data Source tab in the same screen.
Now you can create methods for the Kendra.
As of the publication date of this document, the VoiceWorx platform supports following Amazon Kendra methods,
ListIndices
Query
Below screenshot shows some of the methods already available.
Method: ListIndices
Parameters: Not required
It returns a list of indexes that are already created in Amazon Kendra.
Below is a sample response format – You should get response like this.
[
{
"CreatedAt": number,
"Edition": "string",
"Id": "string",
"Name": "string",
"Status": "string",
"UpdatedAt": number
}
]
Method: Query
Parameters:
Response: Below is a response syntax and you should get response like this. In sample shown below, “string”, number etc should be replaced by the actual value returned by Kendra.
{
"FacetResults": [
{
"DocumentAttributeKey": "string",
"DocumentAttributeValueCountPairs": [
{
"Count": number,
"DocumentAttributeValue": {
"DateValue": number,
"LongValue": number,
"StringListValue": [ "string" ],
"StringValue": "string"
}
}
]
}
],
"QueryId": "string",
"ResultItems": [
{
"AdditionalAttributes": [
{
"Key": "string",
"Value": {
"TextWithHighlightsValue": {
"Highlights": [
{
"BeginOffset": number,
"EndOffset": number,
"TopAnswer": boolean
}
],
"Text": "string"
}
},
"ValueType": "string"
}
],
"DocumentAttributes": [
{
"Key": "string",
"Value": {
"DateValue": number,
"LongValue": number,
"StringListValue": [ "string" ],
"StringValue": "string"
}
}
],
"DocumentExcerpt": {
"Highlights": [
{
"BeginOffset": number,
"EndOffset": number,
"TopAnswer": boolean
}
],
"Text": "string"
},
"DocumentId": "string",
"DocumentTitle": {
"Highlights": [
{
"BeginOffset": number,
"EndOffset": number,
"TopAnswer": boolean
}
],
"Text": "string"
},
"DocumentURI": "string",
"Id": "string",
"Type": "string"
}
],
"TotalNumberOfResults": number
}
Last updated