Indexing

Indexes are reusable collections that help you organize files based on common characteristics or use cases. They provide a systematic way to group related data to annotate.

Overview

Indexes solve the challenge of organizing unstructured data by creating meaningful, task-specific groupings that can be reused across multiple projects. Instead of manually sorting through files or maintaining spreadsheets, indexes automatically maintain collections based on your defined criteria.

Creating an Index

There are two main ways to create and add files to an index:

Before doing either of these: Navigate to “Indexes” and create a new index. This will be where you drop relevant files.

Manual Selection from Catalog

  1. Navigate to the Catalog tab
  2. Browse your files using the integrated folder system
  3. Select the files you want to include
  4. Click “Add to Index” in the top menu
  5. Select the index you wish to add the files to and click “Save”

Creating from Search Results

  1. Use the search interface to find relevant files
  2. Enter your search query or apply filters
  3. Review the search results
  4. Click “Save as Index” in the results view
  5. Select your chosen Index and click “Save”

Key Features

  • Dynamic Grouping: Create flexible collections based on:

    • Semantic similarity
    • Metadata filters
    • Custom tags
    • Search results
  • Deduplication: Automatic detection and removal of duplicate files:

    • Ensures each file appears only once per index
    • Maintains data integrity across projects
    • Prevents redundant processing
  • Traceability: Complete visibility into data usage and lineage:

    • Track which files are used in which workflows
    • Maintain clear lineage of data usage
    • Monitor index changes over time

Best Practices

  1. Start Specific

    • Create focused indexes for specific tasks
    • Use clear naming conventions
    • Document index purpose and criteria
  2. Monitor Usage

    • Track which projects use which indexes
    • Review index contents periodically
    • Update criteria as needs change
  3. Maintain Organization

    • Use consistent naming patterns
    • Document index purposes
    • Archive unused indexes

Indexes are a powerful way to maintain organization as your data grows. Take time to plan your index structure to align with your workflow needs.