Detect and Highlight Entities from Unstructured Text with Pre-trained Models
- How to call and test Business Entity Recognition
- How to access and use the Business Entity Recognition Swagger UI (User Interface)
- How to use the Business Entity Recognition pre-trained models to extract entities from unstructured text
The core functionality of Business Entity Recognition is to automatically detect and highlight any given type of named entity in unstructured text and classify it in accordance with predefined categories. When you finish this tutorial, you will get named entity predictions for the texts you submit to the service.
Business Entity Recognition is a customizable machine learning service, but you can also use the following pre-trained machine learning models:
sap_address_entity
sap_email_business_entity
sap_generic_entities
sap_invoice_header
See Extracted Entities by Pre-trained Model and Free Tier Option Technical Constraints.
This tutorial gives you an introduction to the service using the free tier option for Business Entity Recognition and the available pre-trained machine learning models. If you want to try out the Business Entity Recognition training endpoints to create your own model, check out the tutorial group Use Custom Machine Learning Models to Process Unstructured Text.
- Step 1
You will use Swagger UI, via any web browser, to call the Business Entity Recognition service APIs. Swagger UI allows developers to effortlessly interact and try out every single operation an API exposes for easy consumption. For more information, see Swagger UI.
In the service key you created for Business Entity Recognition in the previous tutorial: Create Service Instance for Business Entity Recognition, you should find (outside the
uaa
section of the service key) an entry calledurl
(as highlighted in the image below).-
To access the Business Entity Recognition Swagger UI, add
/api/v1
to theurl
value, paste it into any web browser and press Enter. -
To be able to use the Swagger UI endpoints, you need to authorize yourself. In the top right corner, click Authorize.
-
Get the
access_token
value created in the previous tutorial: Get OAuth Access Token for Business Entity Recognition Using Any Web Browser, then add bearer in front of it, and enter in the Value field.bearer <access_token>
-
Click Authorize and then click Close.
Choose the correct value that needs to be entered when authorizing the Swagger UI for the Business Entity Recognition API.
-
- Step 2
Use the POST /inference/jobs endpoint to submit the text to be extracted and choose the pre-trained machine learning model you want to use to extract entities from the text.
Click the endpoint name to expand it.
Click Try it out.
In payload, enter the
text
you want to extract named entities from,modelName
, andmodelVersion
. In thissap_address_entity
pre-trained model example, you can use the following:JSONCopy{ "text":"Warehouse 02 (UK) Ltd c/o Excel Logistics Waindyke Way WF6 1TF. West Yorkshire United Kingdom.", "modelName":"sap_address_entity", "modelVersion":2 }
Click Execute.
Copy the
id
from the Response body to see the result of the extraction in the next step.
- Step 3
Use the GET /inference/jobs/{
jobId
} endpoint to see the text extraction results and the confidence level of thesap_address_entity
model.-
Click the endpoint name to expand it.
-
Click Try it out.
-
Enter the
jobId
obtained in the previous step and click Execute.
You should receive a response like the following:
In the response, you will find the prediction for the extracted entities. The prediction is made with a probability indicated by the confidence field which represents how certain the model is about the value provided. A confidence of 1 means that the model is 100% sure about its prediction.
You have now successfully used the
sap_address_entity
model to get text entity predictions for the unstructured text you submitted to Business Entity Recognition.You can also use the following payload example with the
sap_address_entity
model:JSONCopy{ "text":"STAG GmbH Beriiner Chaussee 29, 39307 Genthin Logistik / Einkauf Tel.: 039 33/821 - 223 Fax: 039 33/ 821 - 259.", "modelName":"sap_address_entity", "modelVersion":2 }
To try out the
sap_address_entity
model with your own unstructured text, keep in mind the model scope and supported languages, as detailed insap_address_entity
and Supported Languages. -
- Step 4
Use the POST /inference/jobs endpoint to submit the text to be extracted and choose the pre-trained machine learning model you want to use to extract entities from the text.
Click the endpoint name to expand it.
Click Try it out.
In payload, enter the
text
you want to extract named entities from,modelName
, andmodelVersion
. In thissap_email_business_entity
pre-trained model example, you can use the following:JSONCopy{ "text":"11577210 - FAGOR EDERLAN MEXICO § Dear colleagues, there any way to identify which procurement division has transmitted 1339846 this payment to ABC Ltd. as there is no identification number. Are you able to help? Thank you: Doe, Jane <jane.doe@doe-company.com>: 15 May 2019 15:25 To: Doe, John <john.doe@doe-company.com>Cc: Smith, John <john.smith@doe-company.com>: /432216153 328,439.24 Thanks and best Jane. Doe, Jane. Accountant, Procurement Department, Doe Company, S. L. P.Phone+52 444 880 2300. 1114 Doe Company<doe-company.com>", "modelName":"sap_email_business_entity", "modelVersion":2 }
Click Execute.
Copy the
id
from the Response body to see the result of the extraction in the next step.
- Step 5
Use the GET /inference/jobs/{
jobId
} endpoint to see the text extraction results and the confidence level of thesap_email_business_entity
model.-
Click the endpoint name to expand it.
-
Click Try it out.
-
Enter the
jobId
obtained in the previous step and click Execute.
You should receive a response like the following:
In the response, you will find the prediction for the extracted entities. The prediction is made with a probability indicated by the confidence field which represents how certain the model is about the value provided. A confidence of 1 means that the model is 100% sure about its prediction.
You have now successfully used the
sap_email_business_entity
model to get text entity predictions for the unstructured text you submitted to Business Entity Recognition.You can also use the following payload example with the
sap_email_business_entity
model:JSONCopy{ "text":"Von: lena nova <office@scomputercenter.bg> Gesendet: Montag, 21. Januar 2019, 19:13 Uhr An: canu, ana <ana.canu@company.com>; office@computercenter-bg.com Betreff: RE: M & B EOOD 2639710315 Sehr geehrter Kunde, ich sende Ihnen Informationen für unsere Zahlung. Proforma 198305906 ist ein Teil der Rechnung 4830476570 / 11.01.2019. Rechnungsdatum Betrag in EUR 4830473772 14.12.2018 9,28 € 4830474632 21.12.2018 29,16 € 4830475244 28.12.2018 46,19 € 4830475919 4.1.2019 9,28 € 198305906 10.1.2019 9,29 € 103,20 € 4830476570 11.1.2019 37,13 198305906 10.1.2019 -9,29 € 27,84 Einen schönen Tag noch! Mit freundlichen Grüßen, lena lena nova Verkaufsleiterin M & B EOOD-Computer Center 11, Angel Kanchev Str; 1000, Sofia, Bulgarien Tel.: + 359 2 981 58 57; Mobil: + 359 889 61 13 21 E-Mail: office@scomputercenter.bg Besuchen Sie uns unter: www.computercenter.bg", "modelName":"sap_email_business_entity", "modelVersion":2 }
To try out the
sap_email_business_entity
model with your own unstructured text, keep in mind the model scope and supported languages, as detailed insap_email_business_entity
and Supported Languages. -
- Step 6
Use the POST /inference/jobs endpoint to submit the text to be extracted and choose the pre-trained machine learning model you want to use to extract entities from the text.
Click the endpoint name to expand it.
Click Try it out.
In payload, enter the
text
you want to extract named entities from,modelName
, andmodelVersion
. In thissap_generic_entities
pre-trained model example, you can use the following:JSONCopy{ "text":"In 1988, 31.6% of blacks lived in poverty, compared with 10.1% for whites and 26.8% for Hispanics.", "modelName":"sap_generic_entities", "modelVersion":1 }
Click Execute.
Copy the
id
from the Response body to see the result of the extraction in the next step.
- Step 7
Use the GET /inference/jobs/{
jobId
} endpoint to see the text extraction results and the confidence level of thesap_generic_entities
model.-
Click the endpoint name to expand it.
-
Click Try it out.
-
Enter the
jobId
obtained in the previous step and click Execute.
You should receive a response like the following:
In the response, you will find the prediction for the extracted entities. The prediction is made with a probability indicated by the confidence field which represents how certain the model is about the value provided. A confidence of 1 means that the model is 100% sure about its prediction.
You have now successfully used the
sap_generic_entities
model to get text entity predictions for the unstructured text you submitted to Business Entity Recognition.You can also use the following payload example with the
sap_generic_entities
model:JSONCopy{ "text":"Sie liegt sechs Kilometer nordöstlich des Stadtzentrums von Pardubice und gehört zum Okres Pardubice.", "modelName":"sap_generic_entities", "modelVersion":1 }
To try out the
sap_generic_entities
model with your own unstructured text, keep in mind the model scope and supported languages, as detailed insap_generic_entities
and Supported Languages. -
- Step 8
Use the POST /inference/jobs endpoint to submit the text to be extracted and choose the pre-trained machine learning model you want to use to extract entities from the text.
Click the endpoint name to expand it.
Click Try it out.
In payload, enter the
text
you want to extract named entities from,modelName
, andmodelVersion
. In thissap_invoice_header
pre-trained model example, you can use the following:JSONCopy{ "text":"-----Original Message----- From: Duma Trunchi, Regina <DumaTrunchi@company.com> Sent: Thursday, December 20, 2018 4:33 PM To: Dubey, Ashish <Ashish.Dubey@company.com> Cc: Grc, Sony <grcsony@eu.company.com> Subject: CLEARING//226.530,63eur//phoxdistri Hi Ashish, Please do the clearing as per the attachment. Thank you, Duma-Trunchi Regina Accounts Receivable Finance company Pvt Ltd F&A services operated by AKAI Pvt Ltd Registered office: The Heights, 116 Glenurquhart Road, BALLAUGH, Surrey. KT27 0XW. UK Registered Company Number: 2522874", "modelName":"sap_invoice_header", "modelVersion":1 }
Click Execute.
Copy the
id
from the Response body to see the result of the extraction in the next step.
- Step 9
Use the GET /inference/jobs/{
jobId
} endpoint to see the text extraction results and the confidence level of thesap_invoice_header
model.-
Click the endpoint name to expand it.
-
Click Try it out.
-
Enter the
jobId
obtained in the previous step and click Execute.
You should receive a response like the following:
You have now successfully used the
sap_invoice_header
model to get text entity predictions for the unstructured text you submitted to Business Entity Recognition.You can also use the following payload example with the
sap_invoice_header
model:JSONCopy{ "text":"Order number 12345678 Order type Sales Force Order abcdef Date 01/01/2001 Customer number 9876543 Customer service 123/ 456 6789 E-mail eshop@gmail.com Your Reference: Order receipt confirmation Customer address 508 W. St Margarets St. Brooklyn, NY 11228 Billing address 508 W. St Margarets St. Brooklyn, NY 11228 Ordered by Fenton Moon Shipping type: UPS 2nd Day Air PM Terms of payment: 30 Days Net due Your message: Item Material Description Qty. Price per Unit Net steel gray / Length 12 mm / 1 35.00 35.00 Gross Amount USD 35.00 Freight/ packaging USD 15.00 Tax 2 % USD 1 Tax 4 % USD 2 Total amount USD 53", "modelName":"sap_invoice_header", "modelVersion":1 }
To try out the
sap_invoice_header
model with your own unstructured text, keep in mind the model scope and supported languages, as detailed insap_invoice_header
and Supported Languages.Congratulations, you have completed this tutorial.
-
- Authorize Swagger UI
- Enter inference text for sap_address_entity
- Get inference response from sap_address_entity
- Enter inference text for sap_email_business_entity
- Get inference response from sap_email_business_entity
- Enter inference text for sap_generic_entities
- Get inference response from sap_generic_entities
- Enter inference text for sap_invoice_header
- Get inference response from sap_invoice_header