This page provides an overview of processing-related APIs.

View provides a host of processing capabilities to prepare and process data for numerous use cases including intelligent data management, analytics, and artificial intelligence. Natively within the View solution are experiences for data discovery and AI-powered conversations with your data.

The following core steps are performed for each data asset processed by View:

  • Type detection - rather than rely on hints, content-types, or extensions, View performs magic number and file signature analysis to accurately detect the type of data
  • Semantic cell extraction - View identifies regions within data assets that have a high probability of semantic correlation and extracts them as cells
  • Semantic chunking - with semantic cells identified, View performs chunking within semantic cell data to balance processing throughput with information fidelity
  • UDR generation - View produces a Universal Data Representation (UDR) metadata asset used to identify features, geometry, and data contents. UDR documents provide the foundation for Lexi search
  • Embeddings generation - embeddings are generated against semantic chunk contents using any model you specify

The resultant processed assets (metadata, graph, vector) are then persisted in queryable repositories used not only by View but also exposed to you for consumption outside of View.