Skip to main content

Event Types

Overview

This guide provides an overview of the different types of events generated by R!AN and details how these events are categorized and stored. Understanding these event types and their destinations is crucial for integrating with the system, allowing for effective data management and retrieval.

Event Categories

R!AN generates three primary types of events:

  1. Index Events
  2. Graph Events
  3. Vector Events

Each event type serves a specific purpose and is stored in a database optimized for its data structure.


1. Index Events

Index Event

Description: Index events are structured data points optimized for quick retrieval and efficient storage. They are designed to facilitate rapid querying and data retrieval operations.

Use Cases:

  • Full-text search
  • Filtering and sorting records
  • Quick access to structured information

Destination Databases: Relational databases like PostgreSQL, MySQL, or specialized search engines like Elasticsearch.


2. Graph Events

Graph Event

Description: Graph events represent relationships between concepts/entities, making them ideal for understanding connections and network structures within your data.

Use Cases:

  • Social network analysis
  • Recommendation systems
  • Knowledge graphs

Destination Databases: Graph databases such as Neo4j , which allow for complex queries over relationships.


3. Vector Events

Vector Event

Description: Vector events contain high-dimensional contextually-aware vectors used for similarity search and machine learning applications. They are essential for tasks involving embeddings and vector space computations.

Use Cases:

  • Semantic search space
  • Recommendation engines
  • Graph Neural Networks

Destination Databases: Vector databases like PGVector, or Embedded Postgres, designed for efficient similarity search and retrieval in large datasets.


Summary

R!AN's event-based architecture allows for efficient data processing and storage by categorizing data into Index, Graph, and Vector events. Each event type is optimized for specific use cases and stored in databases suited to their data structures, enabling high-performance data retrieval and analysis.

By understanding these event types and their schemas, you can effectively integrate R!AN into your data processing workflows and leverage its capabilities for advanced data management.