Async analysis for topic modeling

To determine the topics in a document set, use the StartTopicsDetectionJob to start an asynchronous job. You can monitor topics in documents written in English or Spanish.

Before you start

Before you start, make sure that you have:

• Input and output buckets—Identify the Amazon S3 buckets that you want to use for input and output. The buckets must be in the same Region as the API that you are calling.

• IAM service role—You must have an IAM service role with permission to access your input and output buckets. For more information, see Role-based permissions required for asynchronous operations.

Using the AWS Command Line Interface

The following example demonstrates using the StartTopicsDetectionJob operation with the AWS CLI

The example is formatted for Unix, Linux, and macOS. For Windows, replace the backslash (\) Unix continuation character at the end of each line with a caret (^).

aws comprehend start-topics-detection-job \
                --number-of-topics topics to return \
                --job-name "job name" \
                --region region \
                --cli-input-json file://path to JSON input file

For the cli-input-json parameter you supply the path to a JSON file that contains the request data, as shown in the following example.

{
    "InputDataConfig": {
        "S3Uri": "s3://input bucket/input path",
        "InputFormat": "ONE_DOC_PER_FILE"
    },
    "OutputDataConfig": {
        "S3Uri": "s3://output bucket/output path"
    },
    "DataAccessRoleArn": "arn:aws:iam::account ID:role/data access role"
}

If the request to start the topic detection job was successful, you will receive the following response:

{
    "JobStatus": "SUBMITTED",
    "JobId": "job ID"
}

Use the ListTopicsDetectionJobs operation to see a list of the topic detection jobs that you have submitted. The list includes information about the input and output locations that you used and the status of each of the detection jobs. The example is formatted for Unix, Linux, and macOS. For Windows, replace the backslash (\) Unix continuation character at the end of each line with a caret (^).

aws comprehend list-topics-detection-jobs \-- region

You will get JSON similar to the following in response:

{
    "TopicsDetectionJobPropertiesList": [
        {
            "InputDataConfig": {
                "S3Uri": "s3://input bucket/input path",
                "InputFormat": "ONE_DOC_PER_LINE"
            },
            "NumberOfTopics": topics to return,
            "JobId": "job ID",
            "JobStatus": "COMPLETED",
            "JobName": "job name",
            "SubmitTime": timestamp,
            "OutputDataConfig": {
                "S3Uri": "s3://output bucket/output path"
            },
            "EndTime": timestamp
        },
        {
            "InputDataConfig": {
                "S3Uri": "s3://input bucket/input path",
                "InputFormat": "ONE_DOC_PER_LINE"
            },
            "NumberOfTopics": topics to return,
            "JobId": "job ID",
            "JobStatus": "RUNNING",
            "JobName": "job name",
            "SubmitTime": timestamp,
            "OutputDataConfig": {
                "S3Uri": "s3://output bucket/output path"
            }
        }
    ]
}

You can use the DescribeTopicsDetectionJob operation to get the status of an existing job. The example is formatted for Unix, Linux, and macOS. For Windows, replace the backslash (\) Unix continuation character at the end of each line with a caret (^).

aws comprehend describe-topics-detection-job --job-id job ID 

You will get the following JSON in response:

{
    "TopicsDetectionJobProperties": {
        "InputDataConfig": {
            "S3Uri": "s3://input bucket/input path",
            "InputFormat": "ONE_DOC_PER_LINE"
        },
        "NumberOfTopics": topics to return,
        "JobId": "job ID",
        "JobStatus": "COMPLETED",
        "JobName": "job name",
        "SubmitTime": timestamp,
        "OutputDataConfig": {
            "S3Uri": "s3://output bucket/ouput path"
        },
        "EndTime": timestamp
    }
}

Using the SDK for Python or AWS SDK for .NET

For SDK examples of how to start a topic modeling job, see Use StartTopicsDetectionJob with an AWS SDK or command line tool.