×
Community Blog Classify and grade data using AI large language models

Classify and grade data using AI large language models

This article introduces using AI large language models to classify and grade data. It automates data sorting, enhances accuracy, and boosts efficiency in information management.

The Data Security Center (DSC) now supports sensitive data classification and grading powered by a converged architecture of "AI Foundation Models + Domain-Specific Expert Models + Traditional Regex Rules." Compared to previous methods that relied solely on regular expressions and keyword matching, this new solution delivers significant improvements in coverage, accuracy, and intelligence.

Key enhancements include:

Upgraded Recognition Capabilities

By integrating the Qwen foundation model with domain-specific expert models, it supports the automatic identification of over 800 data types. This covers both structured data (e.g., database fields) and unstructured data (e.g., documents, images, and logs).

Improved Accuracy and Recall

It overcomes the limitations of traditional rule-based methods when handling implicit semantics, varying formats, or context-sensitive content, significantly boosting both precision and recall rates.

Flexible Configuration & High-Efficiency Response

Supporting customizable classification and grading strategies, the system provides millisecond-level inference responses to meet compliance and governance requirements across diverse business scenarios.

Seamless Integration & Deployment

Built on a cloud-native architecture, it enables one-click integration into existing data security frameworks without infrastructure modifications, allowing businesses to quickly activate intelligent classification and grading capabilities.

The release of this capability marks a pivotal shift in sensitive data identification from a "rule-driven" to an "AI-driven" era, providing core support for enterprises to build precise, highly efficient, and scalable data security protection systems.

Appendix: Supported Recognition Models via LLM Calls

The following is a list of recognition models that support calls from Large Language Models (LLMs):

Address
Address (Malaysia)
Address (English)
Address (Mainland China)
Residential Address
Name
Name (Malaysia)
Name (English)
Name (Traditional Chinese)
Name (Simplified Chinese)
Personal Name
ID / Document
Passport Number (Mainland China)
US Social Security Number (SSN)
ID Card Number (Hong Kong, China)
Passport
National ID Car
Contact Information
Landline Phone Number (US)
Landline Phone Number (Mainland China)
Personal Phone Number
Banking / Payment
Credit Card Number
Bank Card Number (Mainland China)
Bank Account
Organization / Enterprise Qualification
Tax Registration Certificate Number
Unified Social Credit Code
Organization Code
Business License Number

🔗:https://www.alibabacloud.com/help/dsc/data-security-center/use-cases/classify-and-grade-data-through-ai

0 1 0
Share on

CloudSecurity

28 posts | 2 followers

You may also like

Comments

CloudSecurity

28 posts | 2 followers

Related Products

  • Security Center

    A unified security management system that identifies, analyzes, and notifies you of security threats in real time

    Learn More
  • Data Security Center (Original SDDP)

    An all-in-one data security solution that provides various features, such as sensitive data detection, classification, grading, and de-identification, to help you meet compliance requirements specified in General Data Protection Regulation (GDPR) and personal information protection

    Learn More
  • Data Security on the Cloud Solution

    This solution helps you easily build a robust data security framework to safeguard your data assets throughout the data security lifecycle with ensured confidentiality, integrity, and availability of your data.

    Learn More
  • Security Solution

    Alibaba Cloud is committed to safeguarding the cloud security for every business.

    Learn More