extra_gated_prompt: Please fill out the following information to verify your identity.
extra_gated_fields:
Name: text
Phone Number: text
Email: text
Organization or School: text
Major: text
Education level: text
Identity:
type: select
options:
- Student
- Teacher
- Employee
extra_gated_heading: Acknowledge license to accept the repository
extra_gated_description: >-
Please fill in the information truthfully and keep your phone line clear.
After verifying your information, you will be able to use our dataset.
extra_gated_button_content: Acknowledge license
task_categories:
- image-to-text
language:
- zh
- en
license: cc-by-nc-nd-4.0
Certainly! Here is the translated version of the invoice annotation dataset description:
Dataset Overview
Name: Invoice Annotation Dataset (IAD)
Overview: This dataset includes thousands of invoice samples from various industries and in different formats. Each invoice has been meticulously annotated by human reviewers, covering almost all important structured information found on invoices such as invoice number, date, vendor name, purchaser details, item descriptions, amounts, tax rates, and total amounts.
Data Sources: Derived from both electronic invoices and scanned paper invoices generated through real commercial transactions, ensuring authenticity and diversity of the data. These invoices span across retail, service, manufacturing, and other sectors, aiding in the adaptability training of models across multiple scenarios.
Annotation Content:
- Text Region Localization: Annotations for each text block's position, facilitating integration with Optical Character Recognition (OCR) technologies.
- Field Classification and Extraction: Classification and extraction of content from key fields on the invoice, such as invoice numbers, dates, amounts, etc.
- Logical Relationship Tagging: Annotating logical relationships between fields, e.g., linking item entries to their corresponding amounts and tax rates.
Application Scenarios: Primarily used for developing and testing automated invoice processing systems, including but not limited to automatic data entry, audit support, compliance checks, etc. Training with this dataset can significantly enhance the accuracy and efficiency of related systems.
Format: The dataset offers outputs in multiple formats, including JSON, XML, etc., making it convenient for integration into various software environments.
License and Access: The dataset follows specific open license agreements, allowing free access for academic research and personal learning purposes. However, commercial use may require authorization or licensing fees.
