Suggesters

This page is for configuring suggesters. A suggester is an in-memory index of terms and short phrases. It can correct for misspellings or recommend completions to common phrases of text within a project.

The following two pipelines use suggester indexes: suggestDocs suggests documents based on their title field bearing likeness to the query parameter q. autocorrect corrects misspellings and can recommend completions.

Example 1 (autocorrect)

Suggesters example

The autocorrect suggester is used by a pipeline with the same name autocorrect. In this example we have pressed the Reindex button of the autocorrect suggester. The autocorrect pipeline takes q as input parameter, and responds with a list of corrections and completions.

Autocorrect example

Example 2 (suggest docs)

Here we have reindexed the suggestDocs suggester; it is used by the suggestDocs pipeline.

The image below shows an execution of the suggestDocs pipeline in the run pipeline page that makes a query for heis and shows a suggested document based on that query.

Suggest docs example