Case Study Details

AI Models Review

Focus

Conduct a comprehensive review and provide strategic recommendations aimed at enhancing AI and natural language processing models for similarity mapping and analysis for text based taxonomies.

Approach

Kuaba led the review of two data science models for the Agency. These models leveraged a combination of machine learning and natural language processing techniques to create mappable relationships between text based taxonomies and their descriptions.

Kuaba undertook a thorough examination of all relevant data, modelling and evaluation methdologies to assess the robustness of existing approaches and identify improvements to realise business objectives.

The review was conducted across a series of consultations, in conjunction with detailed analysis of associated datasets, model architectures and results.

Outcome

  • Improved data analysis to inform business decisions
  • Clear understanding of the strengths and weaknesses of models
  • Actionable, complexity graded recommendations to adapt and enhance the current models
  • General recommendations to improve the standard of all models

Thank you for all of your work. The report is really great. We would like to reference the work you have done in some of our next release communications.

Project information

  • Sector Government
  • Jurisdiction Commonwealth
  • Technology Stack
    • Machine Learning
    • Natural Language Processing
    • Artificial Intelligence