Questions tagged with Amazon Kendra

The QueryResultItem includes a ScoreAttributes of VERY_HIGH, HIGH, MEDIUM and LOW. Is the actual confidence result available, or an indication of the threshold of these values?

I have a scanned PDF with handwritten text and I am trying to search through it using AWS Kendra. However, Kendra doesn't search through the handwritten text. I tried using textract and comprehend, but there seem to be limitations to PDF size and number of pages with both services. Is there a way to allow Kendra to search through this handwritten text?

I need to extract text from some .PDFs documents. The formatting of these texts varies (column where it starts and ends) but the beginning and end are always similar, it starts with the word **DECRETO** and ends with the word **CHEFE + something** In this link you can see an example of the original text: https://i.postimg.cc/Xvk0TXJ9/texto.png Is it possible to do this with AWS tools? What's the best way? **OBS: Text language is PT-BR**

We are trying to execute kendra query API, but its return error like this, "Output": { "__type": "com.amazon.coral.service#UnknownOperationException" }, "Version": "1.0" Please help us to solve this. POST: https://kendra.us-east-1.amazonaws.com Request Body : { "IndexId": "xxxx-xxxx-xxxx-xxxx", "QueryText": "string", "QueryResultTypeFilter": "QUESTION_ANSWER" }

I've a spanish Lex v2 bot and AMAZON.KendraSearchIntent is only available in english. There is any other way to integrate this bot with Kendra? Thanks

When using the quick start at the link below, I receive an error message when I use the testing phrase provided in the documentation. What could be causing this error? Quick start: https://aws-quickstart.github.io/quickstart-quantiphi-lex-kendra-backend/ Kendra intent testing phrase: "What preventive measures can I take" Error: ``` An error has occurred: Invalid Bot Configuration: Amazon Lex could not access your Amazon Kendra index. There is a problem with your configuration. Check the configuration and try your request again. ```

I am trying to write a Lambda function to filter a SearchKendra intent on Lex v2. I would like to filter based on 'accountId', which I have assigned to a document in the relevant bucket by including a .metadata file in the same location. Here is my current Lambda function followed by the error that I am receiving. Any ideas what might be going on? Lambda: ``` def lambda_handler(event, context): kendraQuery = { "AttributeFilter": { "EqualsTo": { "Key": "accountId", "Value": { "StringValue": "123testid" } } }, "IndexId": "sample-faq-index" } return { "dialogAction": { "type": "Delegate", "kendraQueryRequestPayload": kendraQuery } } ``` Here is the error message in the Lex console when testing: `Invalid Lambda Response: Received invalid response from Lambda: Can not construct instance of IntentResponse, problem: The validated object is null at [Source: {"dialogAction": {"type": "Delegate", "kendraQueryRequestPayload": {"AttributeFilter": {"equalsTo": {"key": "accountId", "value": {"stringValue": "123testid"}}}, "IndexId": "sample-faq-index"}}}; line: 1, column: 194] `

When you upload FAQ documents, can you use markup within the Answer column in order to control how answers are formatted when displayed as part of a Lex Response? For example, is there a way to include a hyperlink within an Answer so that it is active in a Response? from: https://docs.aws.amazon.com/lex/latest/dg/faq-bot-kendra-search.html I found a FAQ question for you: ((x-amz-lex:kendra-search-response-question_answer-question-1)) and the answer is ((x-amz-lex:kendra-search-response-question_answer-answer-1)). Answer: Just click on the link. If I'm using the example Response given here with the example answer given here, is there a way to enter 'link' in the Answer so that it will be active when displayed as a Response within Lex?

When uploading FAQ to Kendra, you can use custom a custom CSV as described here: https://docs.aws.amazon.com/kendra/latest/dg/in-creating-faq.html How can you access the information in the custom columns to be displayed as part of a Response in Lex? The examples in *To create an AMAZON.KendraSearchIntent intent and response message* at this link below do not describe this. For example, from basic csvs you can access the result question and answer as part of the a Response in Lex: ``` I found a FAQ question for you: ((x-amz-lex:kendra-search-response-question_answer-question-1)) and the answer is ((x-amz-lex:kendra-search-response-question_answer-answer-1)). ``` https://docs.aws.amazon.com/lex/latest/dg/faq-bot-kendra-search.html

When you upload FAQ to Kendra, there are three values: Question, Answer, and URL. However, the documentation on Lex Responses only shows how to return the Question and the Answer: from: [https://docs.aws.amazon.com/lex/latest/dg/faq-bot-kendra-search.html]() ``` I found a FAQ question for you: ((x-amz-lex:kendra-search-response-question_answer-question-1)) and the answer is ((x-amz-lex:kendra-search-response-question_answer-answer-1)). ``` How do I access the URL associated with question-1 and answer-1? A similar questions was asked at the link below - but incorrectly answered. [https://forums.aws.amazon.com/thread.jspa?threadID=336412]()

Hi! Anyone knows if it is possible to edit a existing custom filed mapping? I would like to change the data type. Thanks!

I want to create a knowledge graph based on a variety of raw documents (essentially technical documentation). Are there any typical patterns or solutions?

I have the following structure in Amazon S3 under my bucket: -/kendra-analysis/ - ead0d0af-e769-4c99-97f7-04e68d1cf8b2.txt -/kendra-analysis-meta/ - ead0d0af-e769-4c99-97f7-04e68d1cf8b2.txt.metadata.json My data source configuration in Amazon Kendra has the following: **Data source bucket :** <Bucket_Name> **Metadata prefix :** kendra-analysis-meta/ **Include prefixes :** kendra-analysis/ It seems like my metadata is not picked up. What am I missing?