Amazon Kendra is an clever search service powered by machine studying (ML). We’re excited to announce the launch of Amazon Kendra Featured Outcomes. This new characteristic makes particular paperwork or content material seem on the high of the search outcomes web page every time a person points a sure question. You need to use Featured Outcomes to enhance the visibility of latest paperwork or to advertise sure paperwork when customers enter sure queries.
For instance, you’ll be able to specify that in case your customers enter the question “new merchandise 2023,” then choose the paperwork titled “What’s new” and “Coming quickly” will characteristic on the high of the search outcomes web page. Moreover, in case your customers incessantly use sure queries, you’ll be able to specify these queries for Featured Outcomes. For instance, should you have a look at your high queries utilizing Amazon Kendra Analytics and discover that particular queries reminiscent of “How does kendra semantically rank outcomes?” and “kendra semantic search” are incessantly used, then it is likely to be helpful for the queries to characteristic the doc titled “Amazon Kendra search 101.”
On this put up, we introduce Featured Outcomes and present you tips on how to use them.
Overview of answer
Featured outcomes allows you to create direct mappings from actual queries to paperwork in your index, permitting you to bypass the same old Amazon Kendra rating course of. Amazon Kendra naturally handles key phrase kind queries to rank essentially the most helpful paperwork within the search outcomes, avoiding extreme that includes of outcomes based mostly on easy key phrases. Featured outcomes are designed for particular queries, relatively than queries which can be too broad in scope. You possibly can experiment with that includes totally different paperwork for various queries, or guarantee sure paperwork get the visibility they deserve.
To observe alongside, it is best to have the next stipulations:
You possibly can skip this step you probably have a preexisting index to make use of for this demo.
Add a pattern dataset to your index
Full the next steps so as to add pattern dataset to your index:
- On the Amazon Kendra console, go to your index and select Information sources.
- Select Add knowledge supply.
- Underneath Accessible knowledge sources, choose Pattern AWS documentation and select Add dataset.
- Enter a reputation in your Information supply identify (reminiscent of
sample-aws-data) and select Add knowledge supply.
Search with out Featured Outcomes
On the Amazon Kendra console, select Search listed content material. Within the question subject, begin with a question reminiscent of “Kendra S3 connectors”.
In search outcomes, “DataSourceConfiguration – Amazon Kendra” is listed as the highest search consequence based mostly on the rating course of. However if you wish to promote “Getting began with an Amazon S3 knowledge supply (Console) – Amazon Kendra,” you’ll be able to bypass the Amazon Kendra rating course of to characteristic this consequence on the high of the search outcomes web page.
Create a Featured Outcomes set
To characteristic sure outcomes, you have to specify an actual match of a full textual content question, not a partial match of a question utilizing a key phrase or phrase contained inside a question. For instance, should you solely specify the question “Kendra” in a featured consequence set, queries reminiscent of “How does Kendra semantically rank outcomes?” is not going to render the Featured Outcomes. For extra info on limits, see Quotas for Amazon Kendra. To create a Featured Outcomes set, full the next steps:
- Within the navigation pane, select Featured outcomes, beneath Enrichments.
- Select Create set.
- Enter a reputation in your set (reminiscent of
kendra_connector_feature) and select Subsequent.
- Enter a key phrase to seek out gadgets to characteristic (
kendra s3 connectors).
- Choose Getting began with an Amazon S3 knowledge supply (Console) – Amazon Kendra from the search outcomes.
- Select Subsequent.
- Select Add question.
- Enter a question string (reminiscent of
kendra s3 connectors) and select Add.
- Select Subsequent.
- On the Evaluation and create web page, select Create.
Your Amazon Kendra index is now prepared for pure language queries.
Search with Featured Outcomes
On the Amazon Kendra console, select Search listed content material. Within the question subject, enter the key phrase used within the characteristic outcomes set
kendra s3 connectors.Now, it is best to see Getting began with an Amazon S3 knowledge supply (Console) – Amazon Kendra featured as the highest consequence on the search web page
For extra details about querying the index, see Querying an Index.
To keep away from incurring future fees and to wash out unused roles and insurance policies, delete the assets you created:
- On the Amazon Kendra index, select Indexes within the navigation pane.
- Choose the index you created and on the Actions menu, select Delete.
- To substantiate deletion, enter
Deletewhen prompted and select Delete.
Wait till you get the affirmation message; the method can take as much as quarter-hour.
On this put up, you realized tips on how to use Amazon Kendra Featured Outcomes to advertise content material in an enterprise search answer.
There are lots of further options that we didn’t cowl. For instance:
- You possibly can allow user-based entry management in your Amazon Kendra index, and limit entry to paperwork based mostly on the entry controls you’ve already configured.
- You possibly can map object attributes to Amazon Kendra index attributes, and allow them for faceting, search, and show within the search outcomes.
- You possibly can rapidly discover info from webpages (HTML tables) utilizing Amazon Kendra tabular search.
To study extra about Amazon Kendra, refer Amazon Kendra Developer Information.
Concerning the Authors
Maran Chandrasekaran is a Senior Options Architect at Amazon Internet Providers, working with our enterprise clients. Exterior of labor, he likes to journey.
Kartik Mittal is a Software program Engineer at Amazon Internet Providers, engaged on Amazon Kendra, an enterprise search engine. Exterior of labor, he enjoys mountaineering and likes to journey.
Surya Ram is a Software program Engineer at Amazon Internet Providers, engaged on Amazon Kendra. Exterior of labor, he enjoys chess, basketball and cricket.