Typesense icon

Typesense

Interact with the Typesense search platform

Actions50

Search → Vector Search

AI-generated

Overview

This node performs a vector search operation on a specified collection within a Typesense instance. It allows users to input vector embeddings for similarity search, specify the vector field to search in, and configure various search parameters such as pagination, field inclusion/exclusion, typo tolerance, ranking, highlighting, faceting, grouping, filtering, caching, and advanced options. This node is beneficial for applications requiring semantic or similarity-based search, such as recommendation systems, content discovery, and AI-driven search functionalities.

Use Case Examples

  1. A user wants to find documents similar to a given vector embedding in a Typesense collection for a recommendation engine.
  2. A semantic search application uses this node to retrieve relevant documents based on vector similarity rather than keyword matching.

Properties

Name Meaning
Collection Name Select a collection from your Typesense instance to perform the vector search.
Vector Query Vector embeddings for similarity search provided as comma-separated numbers.
Vector Field Select the vector field within the collection to search in.
Return All Boolean flag to return all results or limit the number of results.
Limit Maximum number of results to retrieve when 'Return All' is false.
Page Page number for pagination of results.
Include Fields Fields to include in the search results. Leave empty to include all fields.
Exclude Fields Fields to exclude from the search results.
Typo Tolerance Options Settings to control typo tolerance behavior during the search, including enabling typo tolerance, number of typos allowed, and related thresholds.
Ranking & Relevance Options to customize ranking and relevance of search results, such as query weights, text match type, prioritization, pinned/hidden hits, synonyms, and exhaustive search.
Highlighting Options to enable and customize highlighting of search results, including fields to highlight, tags, and snippet thresholds.
Faceting Options Settings for faceting strategy, maximum facet values, facet queries, and sampling.
Grouping Options Options to control grouping behavior, such as grouping missing values.
Filter Options Settings to enable lazy filtering and limit filter candidates.
Caching Options to enable caching of search results and set cache TTL (time to live) in seconds.
Advanced Options Additional advanced search options including presets, stopwords, validation, conversation context, limits, and search cutoff time.

Output

JSON

  • results - Array of search results matching the vector query with fields as specified by include/exclude options.

Dependencies

  • Typesense API key credential for authentication with the Typesense instance.

Troubleshooting

  • Ensure the selected collection exists and is accessible in the Typesense instance to avoid collection retrieval errors.
  • Verify that the vector query is correctly formatted as comma-separated numbers to prevent query parsing errors.
  • Check that the vector field specified exists in the selected collection to avoid field-related errors.
  • If the node throws errors related to unsupported resources or operations, confirm that the resource is 'Search' and operation is 'Vector Search' as supported by the node.
  • Network or authentication issues with the Typesense instance may cause API call failures; verify credentials and network connectivity.

Discussion