I&T Solution

Reference No.

S-0254

Solution Name

FormExtractor

Solution Description

FormExtractor is a versatile service that uses artificial intelligence (AI) and optical character recognition (OCR) technologies to automatically capture, index, and extract data from physical forms and documents and convert them into structured digitalised data that can then be subsequently processed by other software, applications, or systems.


FormExtractor is designed to help automate business processes and streamline customer experiences. FormExtractor currently has native support for the data extraction of these documents: business registration, food, travel agency, and hawker licenses in Hong Kong; identity card (ID) information in Hong Kong; and general shopping receipts.


Application Areas

City Management

Commerce and Industry

Development

Education

Employment and Labour

Environment

Finance

Food

Health

Housing

Infrastructure

Population

Recreation and Culture

Social Welfare

Transport

Technologies Used

Artificial Intelligence (AI)

Cloud Computing

Machine Learning

OCR

Use Case

FormExtractor helped expedite the processing of funding applications from Li Ka Shing Foundation's (LKSF) “Instant Relief Fund”' programme. We partnered with Google to use Vision API to extract business registration numbers and company names from physical documents. We also worked with Wilson Parking to automate the process of scanning shopping receipts for their customers.

Here are some of the other use cases of FormExtractor:

- Automate data capture and data entry tasks when rolling out customer loyalty programmes (e.g., real estate, retail/commerce, airline, or other B2C firms)

- Standardise how data is digitalised

- Simplify identification processes (i.e., for accessing banking and insurance products and services)

- Scan legal documents and extract important data points and clauses (i.e., in financial services)

- Detect and mitigate risks to fraud or other compliance issues by accurately cross-checking user inputs against structured data from scanned receipts or documentation (e.g., insurance claims, banking services)

- Streamline paperwork, maintain business continuity, and enhance the efficiency of teams under work-from-home schedules or arrangements by enabling data from physical documents to be submitted remotely and online

- Significantly reduce manual work and human error in data entry and paperwork

- Improve data quality and considerably decrease errors and misinterpretation


If any government department would like to conduct PoC trial or technology testing on the I&T solution, please contact Smart LAB.