1. @sitecore-cloudsdk/search/server

SearchPersonalizationOptions

Version:

Type

Type

Import path

@sitecore-cloudsdk/search/server

Options for personalizing search results.

Signature

export type SearchPersonalizationOptions = { attributes: ArrayOfAtLeastOne<string> } & (
  | { algorithm: 'affinity' }
  | { algorithm: 'mlt'; ids: ArrayOfAtLeastOne<string> }
);

Properties

Name

Type

Description

Value

attributes

string[]

Required.

List of attributes to use for determining similar content. The attributes must be enabled for personalization in Sitecore Search.

To check that the attribute is enabled for personalization in Sitecore Search, click Administration > Domain Settings > Attributes > the attribute of your choice > Use For Features > Personalization.

Minimum length: 1.

  • "brand"

  • "color"

algorithm

string

Required.

Algorithm to use to retrieve personalized results for the site visitor.

Set this value either to "mlt" or "affinity". These two algorithms are similar but use different techniques to personalize results:

  • "mlt" - uses specific index documents from the site visitor's past interactions and shows similar items in the results.

    For example, if the site visitor has interacted with blogs and cloud services, this algorithm will show items that are like blogs about cloud services.

  • "affinity" - does not use specific index documents. Instead, this algorithm considers the site visitor's browsing history and looks at metrics for different attributes.

    For example, this algorithm might learn that a specific visitor likes the color red based on their past searches and clicks. The algorithm then boosts results that have the attribute red.

  • "mlt"

  • "affinity"

algorithm.ids

string[]

List of index document IDs to retrieve similar items for.

For example, the ID of a blog post the visitor has recently clicked.

If unset, Sitecore Search uses past event data for the site visitor.

Only specify ids if algorithm is set to "mlt".

["id1", "id2"]

If you have suggestions for improving this article, let us know!