AI Search - Recently Updated Articles have lower relevancy scores ?

[Anurag Mishra2]

We have been noticing that updating the articles, are reducing the relevancy scores for the article

[sapan]

AI Search displays the most relevant search results for a query first. Machine learning automatically tunes search result relevancy scoring for search experiences based on aggregated user interactions.

Machine learning relevancy is automatically enabled and isn't configurable.

Relevancy models and scoring

AI Search uses a relevancy model to compute a relevancy score for each result returned by a search. Documents with higher relevancy scores appear first in the result set. A result's relevancy score is specific to the particular document, search terms, and user associated with the query.

Each search profile includes its own relevancy model. You can't view, modify, or delete this relevancy model.

Note:AI Search doesn't apply relevancy ranking to *** universal wildcard queries. Results from *** queries appear in an unspecified order.

Search signals and machine learning relevancy tuning

AI Search UX components record signals associated with user searches. These search signals include data on how search users interact with the search input field, auto-complete suggestions, facet and navigation tab filters, Genius Result answer cards, and search results.

Machine learning relevancy uses data from these search signals to intelligently tune relevancy models on a continual basis. Every 30 days, AI Search computes a new version of each relevancy model, iteratively modifying its parameters and regression testing it against aggregated search signal data.

When this tuning process is complete, AI Search compares the existing and new relevancy models to see which one produces better matches for user search behavior as recorded in signal data. It selects the better-performing relevancy model, which remains in use until the next tuning cycle begins.

These relevancy model tuning processes occur separately for each search profile. Changes made to the relevancy model in one search profile don't affect relevancy models in other search profiles.

[ShivarajG]

Hi @sapan,

Each search profile includes its own relevancy model. You can't view, modify, or delete this relevancy model.

Changes made to the relevancy model in one search profile don't affect relevancy models in other search profiles.

The two lines mentioned above appear to have conflicting information regarding the relevancy models in search profiles. It is important to note that each search profile has its own unique relevancy model. This means that changes made to the relevancy model of one search profile do not have any impact on the relevancy models of other search profiles within the system. While you cannot directly view, modify, or delete the relevancy model itself, you can make adjustments to the configuration settings and parameters that influence the behavior of the model. These adjustments can include factors such as weighting and enabling/disabling specific features. Therefore, while you cannot directly access the relevancy model, you have some control over its behavior through the configuration settings.