info@habiledata.com
Read Disclaimer
⟨

Annotating Text from News Articles to Enhance the Performance of an AI Model

Text Annotation Apr 20th, 2024
Annotating Text from News Articles to Enhance the Performance of an AI Model
Scroll Down

Client Profile

The Company: German Construction Technology Company

Industry: Real Estate

Company Headquarters: Germany

The Objective:

Capture, validate and verify information on upcoming or existing construction projects from multi-lingual and multi-format online publications across Europe and USA. Append missing information and manually annotate the 20% data which couldn’t be auto classified.

The Challenges:

  • Identifying, understanding contextually, and tagging project size, phase, location, owner, architects, start date and end date, etc.
  • Handling information from hundreds of articles in a defined tagging lifecycle of 24 hours while adhering to quality standards
  • Onboarding, training, and managing team of data annotators experienced at validating auto-classified architectural information

HabileData’s Solution

HabileData’s data specialists defined and documented KPIs, SOPs and other metrics based on project assessment and understanding of business needs. It was used to develop and deploy a robust validation, verification and data append process to ensure accuracy and credibility of annotation.

  • Tagged property type, project start/end date, site location, size, cost, and construction phases, etc.
  • Data was plugged in real-time on the client portal through secure credentials.
  • Manual validation and verification checked the accuracy of auto-tagged data.
  • Missing information was appended to improve the overall performance of the AI model.
  • Two-step quality check process for each batch of articles ensured labelling accuracy.

Business Impact

  • Accurate text annotations improved the performance of the AI model
  • Text annotation lifecycle reduced from days to just few hours
  • Precise AI algorithm improved customer acquisition rate
  • Up to 50% cost savings for the client through offshoring model

Value Addition

Automated classification and validation followed with accurate annotation of text from thousands of news articles helped to scale the AI model for German construction technology company.

Go to Top