Commercial General Liability Section OCR

 

This article describes how to build an OCR API that extracts data from the Commercial General Liability Section using our deep learning engine. If you want to automate your workflow, this article is for you. 

 

 

Prerequisites

  1. You’ll need a free beta account. Sign up and confirm your email to login.
  2. You’ll need at least 20 Commercial General Liability Section images or pdfs to train your OCR.

 

 

Define your Commercial General Liability Section use case

 

First, we’re going to define what fields we want to extract from your Commercial General Liability Section. 

 

 

Commercial General Liability Section OCR API

Commercial General Liability Section OCR API

Commercial General Liability Section OCR API

 

General Information

  • Agency
  • Carrier
  • Agency Customer ID
  • Policy Number
  • Effective Date
  • Applicant
  • NAIC Code
  • Date

 

Coverages

  • Property Damage Deductibles
  • Bodily Injury Deductibles per claim
  • Bodily Injury Deductibles per occurrence

 

Limits

  • General Aggregate Limit
  • Products & Completed Operations Aggregate
  • Personal & Advertising Injury
  • Each Occurrence
  • Damage to Rented Premises (each occurrence)
  • Medical Expense (any other person)
  • Employee Benefit

 

Premium

  • Premises Operations Premium
  • Products Premium
  • Other Premium
  • Total Premium

 

Schedule of Hazards

  • Loc #
  • Haz #
  • Classification
  • Class Code
  • Premium basis
  • Exposure
  • Terr
  • Premops Rate

 

Claims Made

  • Proposed Retroactive Date
  • Entry Date

 

Employee Benefits Liability

  • Deductible per Claim
  • Number of Employees
  • Number of Employees covered
  • Retroactive Date

 

​​​​​​​Contractors

  • Paid to Subcontractor
  • % of work subcontracted
  • Number of Full Time Staff
  • Number of Part-Time Staff

 

​​​​​​​Products / Completed Operations

  • Product
  • Annual Gross Sales
  • Number of Units
  • Time to Market
  • Expected Life
  • Intended Use
  • Principal Component

 

​​​​​​​Additional Interest / Certificate Recipient

  • Name
  • Address 
  • Rank
  • Reference

 

That’s it for our use case. Feel free to add any other relevant data to fit your requirements. 

 

 

Deploy your API

 

Once you have defined the list of fields you want to extract from your Commercial General Liability Section, head over to the platform and press the ‘build a new endpoint’ button.

 

You land now on the setup page. Here is the image you can use for setting up the API, and my set up looks like this:

 

 

Set up your  Commercial General Liability Section OCR API

 

Once you’re ready, click on the “next step” button. We are going to specify the data types for each of the fields we want our API to extract.


 

To go further, you can download this json config to set up your data model or do it manually.

 

General Information

  • Agency: type String that never contains numeric characters. 
  • Carrier: type String that never contains numeric characters. 
  • Agency Customer ID: type String without specifications. 
  • Policy Number: type String without specifications. 
  • Effective Date: type Date with US format. 
  • Applicant: type String that never contains numeric characters. 
  • NAIC Code: type Number without specifications. 
  • Date: type Date with US format. 

 

Coverages

  • Property Damage Deductibles: type Number without specifications. 
  • Bodily Injury Deductibles per claim: type Number without specifications. 
  • Bodily Injury Deductibles per occurrence: type Number without specifications. 

 

Limits​​​​​​​

  • General Aggregate Limit: type Number without specifications. 
  • Products & Completed Operations Aggregate: type String without specifications.
  • Personal & Advertising Injury: type String without specifications.
  • Each Occurrence: type String without specifications.
  • Damage to Rented Premises (each occurrence): type String without specifications.
  • Medical Expense (any other person): type String without specifications.
  • Employee Benefit: type String without specifications.

 

​​​​​​​Premium

  • Premises Operations Premium: type Number without specifications. 
  • Products Premium: type Number without specifications. 
  • Other Premium: type Number without specifications. 
  • Total Premium: type Number without specifications. 

 

​​​​​​​Schedule of Hazards

  • Loc #: type Number without specifications. 
  • Haz #: type Number without specifications. 
  • Classification: type String that never contains numeric characters. 
  • Class Code: type Number without specifications. 
  • Premium basis: type String that never contains numeric characters. 
  • Exposure: type Number without specifications. 
  • Terr: type Number without specifications. 
  • Premops Rate: type Number that is always an integer. 

 

​​​​​​​Claims Made

  • Proposed Retroactive Date: type Date with US format. 
  • Entry Date: type Date with US format. 

 

​​​​​​​Employee Benefits Liability

  • Deductible per Claim: type Number without specifications. 
  • Number of Employees: type Number without specifications. 
  • Number of Employees covered: type Number without specifications. 
  • Retroactive Date: type Date with US format. 

 

​​​​​​​Contractors

  • Paid to Subcontractor: type Date with US format. 
  • % of work subcontracted: type Number that is always an integer. 
  • Number of Full Time Staff: type Date with US format. 
  • Number of Part-Time Staff: type Date with US format. 

 

​​​​​​​Products / Completed Operations

  • Product: type String that never contains numeric characters. 
  • Annual Gross Sales: type Number without specifications. 
  • Number of Units: type Number without specifications. 
  • Time to Market: type Number without specifications. 
  • Expected Life: type Number without specifications. 
  • Intended Use: type String that never contains numeric characters. 
  • Principal Component: type String that never contains numeric characters. 

 

​​​​​​​Additional Interest / Certificate Recipient

  • Name: type String that never contains numeric characters. 
  • Address: type String without specifications. 
  • Rank: type Number without specifications. 
  • Reference: type String without specifications. 

 

 

Once you’re done setting up your data model, press the Data model ready? start training button at the top of the screen.

 

 

Train your Commercial General Liability Section OCR

 

 

Deep learning Commercial General Liability Section OCR API

 

 

You’re all set! 

 

Now is the time to train your Commercial General Liability Section deep learning model in the Training section of our API. This use case is quite well templated and you should get very high performances with the first training (20 data annotated) :)

 

In a few hours (minutes if you're fast), you’ll get your first model trained and will be able to use your custom OCR API for parsing the Commercial General Liability Section in your application.

 

To get more information about the training phase, please refer to the getting started tutorial. If you have any questions regarding your use case, feel free to reach out!