CAIRE is a web-based application designed to help you organize, cluster, and review AI use cases efficiently. By categorizing AI projects based on their descriptions and titles, CAIRE provides a comprehensive and user-friendly interface to explore and analyze similar projects.
- CSV Upload: Easily upload a CSV file containing your AI project data.
- Project Categorization: Automatically categorize projects based on their descriptions and titles using advanced clustering algorithms.
- Levenshtein Similarity: Identify and highlight projects with similar titles using the Levenshtein distance algorithm.
- Dynamic Visualization: View projects in a user-friendly interface, with font sizes indicating the size of the project clusters.
- Detailed Project Insights: Click on individual projects to see detailed descriptions and find related projects.
-
Navigate to the landing page.
-
Upload your CSV file.
-
Select the columns containing the project titles and descriptions.
-
View the categorized projects in the list.
-
Click on a project title to see its detailed description and view related projects.
Additional Details:
- To test it I used the raw data from the previous AI Inventory Use Case found in: https://github.com/thoppe/Federal-AI-inventory-analysis-2023
Pending:
- Adding environments
- Possibly containarizing it
- Adding other ways to cluster and find similarities
- Extending to excel files