Amazon SageMaker

Input

• This solution works with outlook messages. When outlook messages of msg file format are parsed, this solution provides a response depending upon the intent of the mail. • Supported content type: application/json. • Sample input file will be converted into application json using following commands:

file_path = ('sample_mail.msg')
input_mail = extract_msg.Message(file_path)

• Sample input:

{
  'email_sender': 'NextLabs Testmail <nextlabs.testmail@mphasis.com>',
  'email_body': 'Hi Team,Please tell me how to download the policy document',
  'email_subject': 'Query'
}

Output

• Content type: application/json. • Sample output:

{
  '0': {
    'Response': 'Hello NextLabs Testmail, The policy document can be downloaded from the website, you would require to login to the e-service portal and click on the policy number and then the download button.',
    'Intents': 'Download_policy'
  }
}

• Cancellation, death claim, download policy, fund switch, maturity claim, payment, survival benefit, update number, sum assured, due date, premium amount, last date and maturity date are the set of intents that solution can identify.

Invoking endpoint

AWS CLI Command

If you are using real time inferencing, please create the endpoint first and then use the following command to invoke it

aws sagemaker-runtime invoke-endpoint --endpoint-name "endpoint-name" --body fileb://input.json --content-type application/json --accept application/json out.json

Substitute the following parameters:

• "endpoint-name" - name of the inference endpoint where the model is deployed
• input.json - input msg file to do the inference on
• application/json - MIME type of the given input image (above)
• out.json - filename where the inference results are written to

Resources

• Sample Input Files • Sample Notebook • Sample Output Files

Mphasis DeepInsights
Sample Input File
Sample Jupyter Notebook