Skip to main content
Indexing makes content searchable by creating structured representations of file contents.

​
What Gets Indexed

Indexing captures:
  • Text content from documents
  • Visual descriptions from images
  • Transcripts from audio and video
  • Structured data from spreadsheets
  • Metadata and file properties

​
Indexing Process

Content is indexed in multiple ways:
  • Full-text indexing for exact matches
  • Semantic indexing for meaning-based search
  • Entity indexing for people, places, organizations
  • Temporal indexing for time-based queries

​
Semantic Understanding

Zark uses semantic understanding to index meaning, not just keywords. This allows finding content by concept even when exact words don’t match. Semantic indexing captures:
  • Topics and themes
  • Relationships between concepts
  • Context and intent
  • Similarity between content